Visualizing Health: Tools for Communication, Understanding, and Engagement PDF Free Download
1 / 122
/122
100%
Visualizing Health: Tools for
Communication, Understanding, and
Engagement
Johanne Irene Andreassen
Supervisor: Ankica Babic
Master’s Thesis
Department of Information Science and Media Studies
University of Bergen
June 2, 2025
2
Acknowledgements
I want to express my sincere gratitude to my supervisor, Ankica Babic, for her invaluable guid-
ance, support, and encouragement throughout this year.
A special thank you to Anders for the great collaboration on digital twins and for always be-
ing willing to help.
I am deeply grateful to my boyfriend, Tord, for his constant support and for surviving life with
me on 27 square meters.
Thanks also to all the members of room 634 for making long days feel shorter, and to the
room’s adopted member, Hannah, for making not just this year better, but my whole time as a
student.
Johanne Irene Andreassen
Bergen, June 2, 2025
Abstract
A multi-audience approach to health communication is essential to ensure that individuals can
understand, engage with, and act upon medical advice. As healthcare becomes increasingly
data-driven and personalized, the need for accessible communication tools grows. This mas-
ter’s thesis investigates the role of visualizations, specifically infographics and dashboards, in
enhancing public understanding and engagement with health-related information.
The thesis is structured around three applied case studies, each investigating the commu-
nicative potential of visual tools in distinct health-related contexts. The first case study exam-
ines how infographics based on national dietary guidelines can support engagement with nu-
tritional advice by presenting information in a visually accessible format. This case highlights
how such visualizations can enhance both comprehension and motivation to adopt healthier
choices. The study comprised two iterations. The first involved the development of three in-
fographic prototypes, which focus group participants (n = 72) subsequently evaluated. The
second iteration incorporated participant feedback to refine two selected prototypes.
The second case study examines how visualizations can enhance comprehension of health-
related data by presenting the topic from a perspective in which health functions as a con-
tributing factor rather than the primary object of investigation. Using data from the Quality
of Life Survey (2023), the analysis explores health, demographics, and individual attitudes as
potential predictors of levels of climate change concern. Infographics and a dashboard were
developed to communicate the findings. The results were presented at a workshop during the
Sustainable Development Goals Conference in Bergen (2025), where perspectives on the use
of visualization of health information and the application of digital twins were collected to
inform the design of the final case study.
The third case study explores the development of a static digital twin derived from clus-
ter analysis of health data obtained from the All of Us Research Program. This work was
conducted in collaboration with master’s student Anders Borkenhagen, whose primary focus
was on clustering analysis. A series of visual prototypes was created to represent five distinct
population clusters. These included human body models and icons highlighting key health fea-
tures, displayed on an infographic and a dashboard. The infographic also adopted personalized
health recommendations.
This research adopts a design science methodology, focusing on user-centered and design
principles specific to infographics and dashboards. The findings contribute to the broader
field of visual communication by demonstrating how design can be employed strategically to
iii
enhance clarity, engagement, and accessibility in the context of public health.
iv Abstract
List of Publications
The findings of this research have been communicated through the following publications:
I. Andreassen, Johanne Irene. Babic, Ankica. Tailoring Infographics of the Norwegian
National Dietary Guidelines,Intelligent Health Systems: From Technology to Data
and Knowledge, doi:10.3233/SHTI250091.
II. Andreassen, Johanne Irene. Babic, Ankica. Harnessing Visualization to Enhance
Digital Twins in Health Applications,Envisioning the Future of Health Informatics
and Digital Health, doi:10.3233/SHTI250409.
III. Andreassen, Johanne Irene. Babic, Ankica. Visualizing the Intersection of Climate
Concerns, Health, Attitudes, and Demographics,In press.
Contents
Acknowledgements i
Abstract ii
1 Introduction 1
1.1 ResearchQuestions............................... 2
1.2 Motivation.................................... 3
1.3 ThesisOutline.................................. 4
2 Background 7
2.1 Infographics................................... 7
2.2 Dashboards ................................... 7
2.3 Heart Health and Lifestyle . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.4 HealthyNutrition ................................ 8
2.5 HealthLiteracy ................................. 8
2.6 The Role of Visualizations in Healthcare . . . . . . . . . . . . . . . . . . . . 9
2.7 Data-DrivenHealthcare............................. 10
2.8 DigitalTwins .................................. 10
2.8.1 DigitalCousins............................. 11
2.8.2 Clustering, Prediction, and Visualization in Digital Twin Prototypes . 12
2.8.3 Digital Twin Frameworks . . . . . . . . . . . . . . . . . . . . . . . 12
2.9 PersonalizedMedicine ............................. 15
2.10 Public Concern About Climate Change in Norway . . . . . . . . . . . . . . . 15
3 Methodology 17
3.1 DesignScience ................................. 17
3.1.1 Iterations ................................ 18
3.1.2 Practices ................................ 18
3.1.3 Guidelines ............................... 18
3.2 Threatstovalidity................................ 20
3.3 CaseStudy ................................... 20
3.4 Analyticalstudy................................. 20
vi CONTENTS
3.5 DesignPrinciples................................ 21
3.5.1 Design Principles for Infographics . . . . . . . . . . . . . . . . . . . 21
3.5.2 Design Principles for Dashboards . . . . . . . . . . . . . . . . . . . 21
3.6 Prototyping ................................... 22
3.7 User-CenteredDesign.............................. 22
3.8 DataCollection ................................. 23
3.8.1 FocusGroup .............................. 23
3.8.2 Survey ................................. 23
3.9 Multi-Audience Communication . . . . . . . . . . . . . . . . . . . . . . . . 23
3.10Collaboration .................................. 24
4 Technologies and Tools 25
4.1 CanvaPro.................................... 25
4.2 Flourish..................................... 25
4.3 SPSS ...................................... 26
4.4 Chatgpt ..................................... 26
4.5 SurveyXact ................................... 26
5 Datasets 27
5.1 Quality of Life Survey (2023) . . . . . . . . . . . . . . . . . . . . . . . . . 27
5.2 AllofUs .................................... 28
6 Case Study 1: Enhancing Public Engagement with Dietary Guidelines Through
Infographics 33
6.1 Introduction................................... 33
6.2 Background................................... 33
6.3 FirstIteration .................................. 34
6.3.1 Prototype evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . 37
6.4 SecondIteration................................. 40
6.4.1 User-Centered Design . . . . . . . . . . . . . . . . . . . . . . . . . 42
6.4.2 Prototyping in the Design Process . . . . . . . . . . . . . . . . . . . 43
6.4.3 Application of GRAPHIC Principles . . . . . . . . . . . . . . . . . . 43
6.5 Limitations ................................... 44
6.6 Conclusion ................................... 45
7 Case Study 2: Visualizing Factors Influencing Climate Change Concern 47
7.1 Introduction................................... 47
7.2 Background................................... 47
7.3 Demographics.................................. 48
7.4 Health...................................... 49
7.5 Attitude ..................................... 50
CONTENTS vii
7.6 KeyFindings .................................. 52
7.7 PhrasingIssue.................................. 52
7.8 Infographics................................... 52
7.8.1 Application of GRAPHIC Principles . . . . . . . . . . . . . . . . . . 56
7.9 Dashboards ................................... 56
7.10Survey...................................... 57
7.11UseCases.................................... 58
7.12Conclusion ................................... 58
8 Case Study 3: Designing Static Digital Twin Prototypes 59
8.1 Introduction................................... 59
8.2 Background................................... 59
8.3 Collaboration .................................. 59
8.4 Prototyping ................................... 62
8.4.1 Visualization of Clusters Using Human Figures . . . . . . . . . . . . 62
8.4.2 Icon Design for Health Features . . . . . . . . . . . . . . . . . . . . 63
8.4.3 Dashboard Development . . . . . . . . . . . . . . . . . . . . . . . . 64
8.4.4 Infographic Development . . . . . . . . . . . . . . . . . . . . . . . 65
8.4.5 Application of GRAPHIC Principles . . . . . . . . . . . . . . . . . . 66
8.5 DesignProcess ................................. 67
8.6 DigitalTwinFrameworks............................ 68
8.6.1 Data-Centric Framework . . . . . . . . . . . . . . . . . . . . . . . . 68
8.6.2 Integration Levels Framework . . . . . . . . . . . . . . . . . . . . . 69
8.6.3 Lifestyle-Focused Archetypes Framework . . . . . . . . . . . . . . . 69
8.7 Limitations ................................... 69
8.8 Discussion.................................... 70
8.9 Conclusion ................................... 70
9 Discussion 73
9.1 Implementation of Design Science Principles . . . . . . . . . . . . . . . . . 73
9.2 Answering Research Questions . . . . . . . . . . . . . . . . . . . . . . . . . 75
10 Conclusions and Future Work 79
10.1Conclusions................................... 79
10.2Limitations ................................... 79
10.3FutureWork................................... 80
10.4FinalRemarks.................................. 80
Appendix 91
List of Figures
3.1 Illustration of the relationship between people, practices, problems, and artifacts. 19
5.1 PubMed...................................... 29
5.2 GoogleScholar.................................. 30
5.3 ScienceDirect. ................................. 31
6.1 A customized version of Figure 3.1 created to illustrate the process of identi-
fying a problem and addressing it through the development of an artifact, in
thiscaseinfographics............................... 34
6.2 Infographic1................................... 35
6.3 Infographic2................................... 36
6.4 Infographic3................................... 37
6.5 Focus group voting distribution Andreassen and Babic (2025a). . . . . . . . . 38
6.6 Infographic 2 after second iteration. . . . . . . . . . . . . . . . . . . . . . . 41
6.7 Infographic 3 after second iteration. . . . . . . . . . . . . . . . . . . . . . . 42
7.1 Age........................................ 49
7.2 Gender. ..................................... 49
7.3 Levelofurbanization............................... 49
7.4 Regionaldistribution............................... 49
7.5 Physical health satisfaction. . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
7.6 Long-termillnesses................................ 50
7.7 Overallhealthrate. ............................... 50
7.8 Determination. ................................. 51
7.9 Lifecontrol.................................... 51
7.10Problemsolving. ................................ 51
7.11 Demographics infographic. . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
7.12Healthinfographic. ............................... 54
7.13Attitudeinfographic. .............................. 55
7.14Firstprototype. ................................. 57
7.15Secondprototype................................. 57
7.16Thirdprototype.................................. 57
7.17Fourthprototype. ................................ 57
LIST OF FIGURES ix
7.18 The design development process. . . . . . . . . . . . . . . . . . . . . . . . . 58
8.1 K=5Cluster................................... 61
8.2 The different mean BMI’s of the clusters visualized through BMI Visualizer
Max Planck Institute for Intelligent Systems (2013)............... 63
8.3 Thechosenicons................................. 64
8.4 Illustration of the development process from initial clusters to static digital
twin visualizations in the dashboard. . . . . . . . . . . . . . . . . . . . . . . 65
8.5 Personalized Infographic. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
8.6 Overview of the cluster structure, dashboard design, and an example of a per-
sonalizedinfographic............................... 68
List of Tables
3.1 The seven guidelines for rigorous design science developed by (Hevner et al.,
2004)....................................... 19
6.1 Focus group participant qualitative feedback on each infographic by age group
Andreassen and Babic (2025a).......................... 39
6.2 Application of GRAPHIC Principles in Infographics in Case Study 1. . . . . 44
7.1 Application of GRAPHIC Principles in Infographics in Case Study 2. . . . . 56
8.1 Silhouette scores for different numbers of clusters K. ............. 61
8.2 Cluster characteristics based on mean values of selected health-related features. 62
8.3 Application of GRAPHIC Principles in Infographics in Case Study 3. . . . . 67
Chapter 1
Introduction
Why visualize data? Robert Grant raises this question in his book Data Visualization: Charts,
Maps, and Interactive Graphics Grant (2018). In response, he explains that the human brain
is naturally inclined to recognize patterns and differences, and that we tend to make sense
of spatial relationships through visual perception. In healthcare, where data can be dense and
challenging to interpret, especially for those without a medical background, visualization plays
a crucial role in facilitating understanding. Graphs, charts, and other visual formats can make
complex information easier to understand and act upon Abudiyab and Alanazi (2022).
In recent years, healthcare systems have increasingly adopted visualization tools as part
of their digital infrastructure. These tools support evidence-based decision-making by help-
ing practitioners and stakeholders explore and interpret data more effectively Abudiyab and
Alanazi (2022). At the same time, the growing availability of detailed health data has created
new opportunities to study human behavior and health patterns on a larger scale (Grossglauser
and Saner, 2014). This development has established a foundation for data-driven healthcare,
where clinicians use data to inform decisions aimed at improving diagnosis, treatment, and
prevention (Amri and Abed, 2023).
One of the most promising concepts to emerge from this data-driven paradigm is the med-
ical digital twin. A medical digital twin is a real-time, virtual representation of a patient, built
using data from sensors and other sources, which aims to mirror and update the state of the
physical body (Sun et al., 2022). Although the potential applications are significant, the con-
cept remains abstract for many. This raises the question of how to communicate digital twins
to the general public, particularly in light of common concerns around data privacy, trust, and
unfamiliarity.
Effective communication of complex medical information requires more than technical
precision. Health literacy plays a vital role in shaping how individuals understand and respond
to information about their well-being. Many misunderstandings in healthcare stem from sub-
tle communication failures, which can have serious consequences for patients Graham and
Brookey (2008). Clear and inclusive visual communication is therefore essential. It enables
multi-audience communication, where individuals with varying levels of knowledge and ex-
perience can understand the relevance and implications of health-related data Mörth (2022).
2 Introduction
As healthcare increasingly moves toward a personalized model, it becomes more critical
that everyone, not only professionals, can engage with and understand their health information.
Visualizations must support this by being both accurate and accessible. This thesis examines
how visual tools can enhance the engagement and understanding of health data for a broader
audience.
To address this challenge, the thesis proposes the use of infographics and dashboards as
two complementary visualization tools. Infographics, which are graphical representations of
information, have been shown to enhance understanding by presenting content in a visually
engaging and simplified format McCrorie et al. (2016). In health research, they are frequently
used to share study findings, translate knowledge into practical guidance, and facilitate com-
munication with non-expert audiences Beecher et al. (2023). Their use in patient education is
also becoming increasingly common, as they can help convey complex medical information in
a way that is easier to understand Hernandez-Sanchez et al. (2021).
Dashboards provide another powerful method for visualizing health-related data. They
integrate multiple visual and graphical elements to present layered information, allowing for
the abstraction and simplification of complex data Bach et al. (2022). In clinical practice,
dashboards are frequently used to monitor and support the treatment of individual patients,
providing real-time access to relevant data. A literature review conducted by Dowding et al.
found that dashboards used in healthcare environments were associated with improvements in
care processes and better patient outcomes Dowding et al. (2015).
The implementation of visualization in this thesis takes place on three levels. First, it ex-
amines how visualizations can enhance user engagement by creating infographics that aim
to make the Norwegian dietary guidelines more interesting and comprehensible. Second, it
explores how visual tools can promote understanding of abstract concepts by developing in-
fographics and a dashboard based on survey data related to concerns about climate change.
Finally, it introduces a dashboard and an accompanying infographic that present a static digi-
tal twin model, constructed through cluster analysis, to demonstrate how such models can be
used to communicate personalized health insights.
1.1 Research Questions
To guide this thesis, five research questions were developed to explore the role of visualization
in health communication. The first question explores how visualization can support the under-
standing of health-related information and promote patient engagement. The second addresses
strategies for making abstract health concepts more accessible to diverse audiences. The third
focuses on identifying the most effective visualization formats for digital twin models. The
fourth investigates how visualizations of digital twins can influence lifestyle choices related to
cardiovascular disease. Finally, the fifth question examines current attitudes towards the visu-
alization of digital twins in healthcare.
1.2 Motivation 3
RQ 1: How can visualization support understanding of health-related information and enhance
patient engagement?
RQ 2: How can abstract health concepts be made more understandable for multi-audience
communication?
RQ 3: Which visualization format is most effective for communicating digital twin models?
RQ 4: How can digital twin visualizations enhance cardiovascular disease lifestyle choices?
RQ 5: What are current attitudes towards the visualization of Digital Twins?
1.2 Motivation
The way health information is communicated has a significant impact on how well it is un-
derstood and acted upon. Visualizations can empower both individuals and researchers to
understand better and explore health data. By presenting complex information in an intuitive
and engaging manner, visualizations can bridge the gap between technical data and practical
understanding.
A personal experience with my grandmother sparked my interest in health visualization.
After a knee check-up, she received a medical report, a long page filled with complicated
terminology that was almost impossible to understand without a medical background. As I
helped her look up unfamiliar terms and interpret the report, I realized how confusing and
overwhelming the information was for someone without specialized knowledge.
This experience made me reflect on how differently the situation could have been handled.
Had the doctor’s report employed more fitting language and incorporated visual elements, my
grandmother might have found it easier to comprehend her condition and the available treat-
ment options. I began to consider whether the information could have been more effectively
conveyed through an infographic that visually highlighted the site of the injury alongside ex-
planatory text. Instead, the dense and technical nature of the written report left her feeling
anxious and uncertain. A well-designed visualization might have offered both clarity and re-
assurance in a moment where understanding was crucial.
Since then, I have been motivated to explore how visualization techniques can enhance the
communication of health information. I am particularly interested in the use of infographics
and dashboards to enhance understanding and better cater to varying health literacy levels
among patients, as well as support informed decision-making for healthcare professionals.
4 Introduction
1.3 Thesis Outline
Chapter 1: Introduction
This chapter introduces the central themes of the thesis, including the visualization of health
data, data-driven and personalized healthcare, and digital twins. It explores how these concepts
can be made more accessible to patients and the general public through the use of infographics
and dashboards. The chapter then outlines the research questions guiding the study, presents
the motivation behind the research, and concludes with an overview of the thesis structure.
Chapter 2: Background
The background chapter provides an introduction to the various visualization techniques em-
ployed and an overview of the relevant literature. It begins with discussions on infograph-
ics and dashboards as visualization tools, followed by explorations of heart health, lifestyle,
healthy nutrition, and health literacy. The chapter further addresses the role of visualizations
in healthcare and the emergence of data-driven and personalized healthcare. It then delves into
digital twins, including the concept of digital cousins, clustering and prediction techniques
in digital twin prototypes, and various digital twin frameworks. The chapter concludes with
insights on how health can be a framing element for public concern about climate change,
alongside demographics and personal attitudes.
Chapter 3: Methodology
This chapter outlines the methodological framework of the research. It begins by elaborating
on the application of design science, including its iterative process, practices, and guiding prin-
ciples. Measures taken to ensure research validity and data integrity are then described. The
chapter proceeds to present the different case studies conducted, with particular attention to the
analytical components of Case Studies 2 and 3. It introduces design principles specific to info-
graphics and dashboards, and describes the prototyping and user-centered design approaches
employed in these contexts. Data collection methods, including focus groups and surveys, are
also discussed. The chapter concludes with a reflection on the concept of multi-audience com-
munication and highlights the collaboration with Anders Borkenhagen Borkenhagen (2025).
Chapter 4: Technologies and Tools
This chapter provides an overview of the technologies and tools used throughout the project.
It explains the different types of software and their respective roles in prototyping, data gath-
ering, and data analysis.
Chapter 5: Datasets
The datasets chapter describes the primary data sources used in the research, including the
Quality of Life Survey (2023) and All of Us datasets. It provides contextual information about
these datasets and their relevance to case studies 2 and 3.
1.3 Thesis Outline 5
Chapter 6: Case Study 1 – Enhancing Public Engagement with Dietary Guidelines
Through Infographics
This chapter presents the first case study, which focuses on improving public engagement with
dietary guidelines through the use of infographic design. It outlines the theoretical background
and describes the applied methodologies, including user-centered design, iterative evaluation
of prototypes, and the implementation of the seven GRAPHIC Principles of Public Health
Infographics. Furthermore, it addresses key limitations encountered during the study and con-
cludes with a summary of the main findings.
Chapter 7: Case Study 2 – Visualizing Factors Influencing Climate Change Concern
Chapter Seven investigates the visualization of public concern about climate change in Nor-
way. It examines how demographics, health status, and individual attitudes affect levels of
concern. The chapter presents both infographics and a dashboard design developed to com-
municate the insights from the analysis. It highlights how the GRAPHIC principles were used
to guide the design. The chapter concludes with a discussion of the design development pro-
cess and the broader implications of the study.
Chapter 8: Case Study 3 – Designing Static Digital Twin Prototypes
This chapter presents the third case study, which concerns the development of a static digi-
tal twin. It outlines how Borkenhagen processed the data Borkenhagen (2025), and describes
how the resulting outputs were visualized. The prototyping process involved the development
of cluster visualizations, a dashboard, and an infographic. The GRAPHIC Principles demon-
strate the design choices of the infographic. The chapter analyzes the application of digital
twin frameworks and concludes with a discussion of limitations and possibilities for the devel-
oped prototypes.
Chapter 9: Discussion
This chapter critically reflects on the application of design science principles across the three
case studies. It also brings together findings from the case studies, survey answers, and back-
ground literature to answer the research questions.
Chapter 10: Conclusions and Future Work
The final chapter summarizes the main conclusions of the thesis and acknowledges the study’s
limitations. It proposes directions for future research, such as integrating real-time data into
digital twin prototypes and adaptive visualization techniques in infographics. It concludes with
final reflections on the work completed.
6 Introduction
Chapter 2
Background
This chapter provides an overview of visualization techniques and literature relevant to the top-
ics of this thesis, including visualization in healthcare, data-driven healthcare, digital twins,
and climate change concerns. It begins by exploring infographics and dashboards, and de-
scribes how they clarify complex health information for diverse audiences. The chapter then
addresses the significance of heart health and lifestyle, with a focus on nutrition. It also ad-
dresses health literacy and emerging healthcare techniques, like data-driven healthcare. It
provides a thorough description of digital twins and explains how they can be developed and
evaluated within various frameworks. It ties together data-driven healthcare, digital twins, and
personalized medicine, and concludes with insights into public concern about climate change.
2.1 Infographics
Information graphics, also known as infographics, combine visual elements with text to present
information or data effectively. This representation of information in a graphical format im-
proves understanding and decision-making McCrorie et al. (2016). Infographics are recog-
nized as practical tools for simplifying complex information and enhancing engagement, par-
ticularly in healthcare communication, where they improve comprehension by making infor-
mation more accessible to diverse audiences Andreassen and Babic (2025a); Martin et al.
(2019); Piil et al. (2023). In health research, they are often used to share findings, translate
knowledge, and improve communication with non-expert audiences Beecher et al. (2023). In-
fographics are being used more frequently for patient education and are effective in helping
patients follow the medication plan they receive from healthcare personnel Hernandez-Sanchez
et al. (2021).
2.2 Dashboards
Dashboards utilize a combination of visuals and graphs to provide layers of abstraction and
simplification of complex data, aiming to offer users a concise and time-efficient overview
8 Background
Bach et al. (2022). In a medical context, dashboards can serve multiple functions. They can be
used to monitor the performance of healthcare organizations, in which case they are referred
to as quality dashboards. Alternatively, they can be designed to support the monitoring and
treatment of individual patients, in which case they are known as clinical dashboards. One
example of a quality dashboard is the HIV-data reporting performance dashboard developed
by Gesicho and Babic Gesicho and Babic (2022). Clinical dashboards function as digital tools
that present clinicians with timely and relevant information to support decision-making in
everyday patient care Dowding et al. (2015). A literature review by Dowding et al. found that
clinical dashboards were associated with improved care processes and better patient outcomes
Dowding et al. (2015).
2.3 Heart Health and Lifestyle
Cardiovascular diseases are the leading cause of death globally, and modifiable lifestyle fac-
tors such as smoking, physical inactivity, poor diet, and excessive alcohol consumption sig-
nificantly increase the risk Barbaresko et al. (2018). Adherence to multiple healthy lifestyle
behaviors simultaneously has been shown to reduce the risk substantially. In a comprehensive
review, Barbaresko et al. found that individuals who adopted several healthy lifestyle habits
had a 66% lower risk of developing cardiovascular disease compared to those who followed
none or only one such behavior. Although the studies in the review used different ways to de-
fine and measure healthy lifestyle habits, the overall finding was the same: the more healthy
behaviors people had, the lower their risk Barbaresko et al. (2018).
2.4 Healthy Nutrition
Among the most significant lifestyle choices influencing heart health outcomes is adopting a
nutritious diet. Throughout the life course, diet and nutrition play a foundational role in sup-
porting physical health and functional well-being World Health Organization and Food and
Agriculture Organization of the United Nations (2003). Nutrition is also increasingly recog-
nized as a modifiable factor in the prevention of chronic diseases World Health Organization
and Food and Agriculture Organization of the United Nations (2003). A growing body of evi-
dence shows that dietary changes can have a significant impact on long-term health outcomes
Andreassen and Babic (2025a); World Health Organization and Food and Agriculture Orga-
nization of the United Nations (2003). This highlights the need for engaging visualizations of
nutritional advice.
2.5 Health Literacy
Health literacy refers to the degree to which people can understand health information Graham
and Brookey (2008). Considering varying levels of health literacy is essential to ensure safe
2.6 The Role of Visualizations in Healthcare 9
and effective communication between patients and healthcare providers. Communication bar-
riers in clinical settings often go unnoticed, yet they can significantly impact patient outcomes.
Graham and Brookey highlight that limited literacy skills are among the strongest predictors
of poor health outcomes, as individuals with low reading fluency tend to have less knowl-
edge about their chronic conditions, experience greater difficulty managing their care, and are
less likely to take preventive health measures. Notably, Graham and Brookey also emphasize
that low health literacy is not limited to individuals with poor literacy skills; even those with
adequate general literacy may face challenges in comprehending and applying health-related
information Graham and Brookey (2008).
2.6 The Role of Visualizations in Healthcare
Presenting numerical facts visually rather than in an abstract format makes the information
more concise, universally accessible, and easy to understand Grant (2018). Healthcare man-
agement systems are increasingly using technology to implement various visualization tech-
niques that support evidence-based medical practice Abudiyab and Alanazi (2022). Abudiyab
and Alanazi highlight five main advantages of health data visualization in healthcare:
Firstly, it significantly enhances patient care by enabling real-time monitoring and in-
formed clinical decision-making. By visualizing key health parameters such as oxygen sat-
uration, heart rate, and blood pressure, healthcare providers can quickly detect abnormalities
and intervene promptly, ultimately improving patient outcomes and overall healthcare quality.
Secondly, health data visualization plays a crucial role in identifying disease trends and
patterns. Recognizing patterns in disease prevalence allows healthcare professionals to inves-
tigate underlying causes and implement targeted interventions. For example, tracking obesity
trends through visual analytics can help raise awareness and promote lifestyle modifications at
both the individual and community levels.
Thirdly, the ability to present complex healthcare data in an accessible format benefits
various audiences. Medical data can be challenging to interpret, especially for those outside
the healthcare field. Using graphs, charts, and other visual representations, the data becomes
more comprehensible, ensuring that stakeholders, whether healthcare providers, policymakers,
or the general public, can make informed decisions based on clear insights.
Fourthly, real-time data visualization accelerates healthcare performance by streamlining
clinical workflows and supporting rapid decision-making in critical situations. By reducing
inefficiencies and improving response times, healthcare organizations can enhance both patient
outcomes and overall efficiency, ultimately contributing to better patient care and a stronger
institutional reputation.
Finally, visualization of health data helps to detect errors and dishonest activities within
healthcare facilities, particularly in areas such as medical billing. By improving transparency
and facilitating better communication among stakeholders, including patients, providers, and
insurers, data visualization strengthens financial integrity and reduces losses resulting from
10 Background
misleading claims.
Together, these five advantages demonstrate the vital role of visualization of health data in
improving healthcare delivery, patient outcomes, and system efficiency Abudiyab and Alanazi
(2022).
2.7 Data-Driven Healthcare
The growing accessibility of rich health data presents opportunities to study human behavior
closely (Grossglauser and Saner, 2014). Data-driven healthcare, which involves using data
for informed decision-making in healthcare to serve patients better, is proposed as the future
of disease detection, treatment, and prevention (Amri and Abed, 2023).
Data visualization, along with machine learning, predictive analytics, and natural language
processing, are methods that enable data-driven healthcare (Amri and Abed, 2023). Although
data-driven healthcare holds great potential, challenges and ethical issues must also be consid-
ered.
It faces significant challenges in terms of privacy and data accuracy (Grossglauser and
Saner, 2014). Privacy concerns surrounding patients’ medical data, including the risks of
wrong treatment decisions and misusage of the data, and data accuracy, as behavioral data
collected from sensors is prone to noise and errors, introducing uncertainties that can impact
decision-making (Grossglauser and Saner, 2014).
Ethical considerations such as informed consent, data ownership, and the risk of discrimi-
nation further complicate the usage of data-driven healthcare, making careful regulation nec-
essary (Amri and Abed, 2023).
2.8 Digital Twins
Medical digital twins, a component of data-driven healthcare, refer to a virtual model or sim-
ulation of a part of, or the whole body, of a patient, designed to replicate real-time health
conditions. This concept can be described in various ways. In the field of precision health,
Sel et al. characterize digital twins as virtual models that replicate the structure, context, and
behavior of human bodies or healthcare systems, continuously updating with real-world data
from their physical counterparts (Sel et al., 2024).
Sun et al. further elaborate on digital twins as dynamic, evolving representations of human
organs, tissues, cells, or micro-environments that adjust in response to incoming data and can
predict potential changes in their corresponding physical counterparts (Sun et al., 2023). Sun
et al. also characterize digital twins as real-time virtual replicas that act as a bridge between
the physical and digital worlds, integrating sensor data to maintain an up-to-date, interactive
model (Sun et al., 2022).
The concept of digital twins originally emerged in engineering, where it was applied to
complex systems such as aircraft and entire cities (Björnsson et al., 2020). As Björnsson et
2.8 Digital Twins 11
al. observe, these virtual models were initially developed to facilitate more efficient and cost-
effective testing and development processes, reducing reliance on real-world experimentation
(Björnsson et al., 2020). In healthcare, digital twins serve a similar function by enabling vir-
tual experimentation with treatment strategies, thus supporting clinical decision-making with-
out exposing patients to unnecessary risk (Fitzgerald et al., 2024). Furthermore, digital twins
can play a crucial role in proactive healthcare by facilitating the early detection of medical con-
ditions through real-time data integration (Randles, 2025). Randles highlights that, when con-
nected to wearable devices and continuous monitoring systems, digital twins enable clinicians
to identify emerging health issues before symptoms become clinically apparent or conditions
worsen. This capability facilitates timely intervention, thereby improving patient outcomes
and reducing overall healthcare costs by minimizing complications and avoiding unnecessary
hospitalizations (Randles, 2025).
Digital twins present unique ethical challenges that require careful attention as their use
in healthcare grows. Banerjee et al. highlight that because these models are built on detailed
and often longitudinal personal data, they cannot be fully anonymized in the same way as
traditional health records. This makes strong data governance essential, not only to protect
privacy but also to prevent misuse and abuse. One emerging concept in this area is digital
dignity, which emphasizes the right of individuals to know how their data is used and to retain
some control over their digital representation. This calls for more flexible and transparent
consent processes that allow participants to stay informed and involved over time. There are
also concerns about how insights from digital twins might be applied, for example, in ways that
could unfairly influence insurance decisions or reinforce biases in healthcare. To address these
risks, access to digital twins should be tightly controlled and regularly monitored, with clear
boundaries established around who can use them and for what purposes. Ethical, Legal, and
Social Implications (ELSI) frameworks can help guide these efforts, ensuring that the benefits
of digital twin research do not come at the cost of individual rights or public trust (Banerjee
et al., 2024).
2.8.1 Digital Cousins
The concept of digital twins is much more straightforward when applied to machines and
engines than to human beings. Engineered systems are typically easier to replicate digitally
than natural systems, such as humans, which contain inherent uncertainties and complexities.
Hodgkinson and Elmouttie argue that in fields such as geology, current geological models and
simulations can, at best, produce a digitized “cousin” rather than a true digital twin Hodgkinson
and Elmouttie (2020). Similarly, because creating a fully functional digital human twin is
highly advanced and technically demanding, this master’s thesis focuses on developing a basic
digital representation. Instead of relying on continuous sensor data, the visualizations are
based on data clustering. As such, the visualization results are closer to a “digital cousin” than
a complete digital twin.
12 Background
2.8.2 Clustering, Prediction, and Visualization in Digital
Twin Prototypes
As part of a Digital Twin Project carried out by master’s students from the University of Bergen
in 2024, William Røise, Carl Oskar Kraft Sahlgaard, and Ida Wergeland Sævareid made no-
table contributions to the development of digital twin concepts in healthcare Røise (2024);
Sahlgaard (2024); Sævareid (2024). Collectively, their theses explore how clinical data can
be used to identify similarities between patients and serve as a foundation for decision support
tools, particularly in the field of hip arthroplasty.
Sahlgaard’s thesis focuses on applying clustering methods to group patients based on clin-
ical similarity. His work explores how such clusters can be used to define digital twins, while
also acknowledging the limitations of available data. To address these constraints, he intro-
duces the concept of “digital cousins” to describe patients who share relevant characteristics
but do not fully qualify as digital twins Sahlgaard (2024).
Røise’s thesis builds on this by exploring how digital twins can be integrated into clinical
pathways to support predictive modeling. He implements basic machine learning pipelines to
demonstrate how digital twins can be initialized using event logs and subsequently updated in
real time. His research highlights the potential of combining structured care processes with
data-driven prediction methods Røise (2024).
Sævareid’s thesis synthesizes the insights from both Sahlgaard’s and Røise’s work, focus-
ing on the effective visualization of digital twin concepts for healthcare professionals. Her
project, "Developing Conceptual Designs for Digital Twins in Arthroplasty," investigates how
to communicate complex clinical data and relationships in a clear and meaningful manner. She
developed conceptual models that enable intuitive navigation of patient similarity and clinical
pathways Sævareid (2024).
These theses demonstrate how clustering, predictive modeling, and visualization can be
integrated to develop digital twin designs, which provided the conceptual foundation for the
approach adopted in Case Study 3 of this thesis.
2.8.3 Digital Twin Frameworks
Data-Centric Framework
Demuth et al. (2025) present a data-centric framework for implementing medical digital twins,
addressing the confusion in existing literature where the term “digital twin” has been applied
inconsistently Demuth et al. (2025). They argue that a single digital representation cannot
satisfy the diverse technical and regulatory requirements inherent in real-world clinical appli-
cations. Their proposed framework includes three main types of digital representations: multi-
modal dashboards to provide perception and documentation aids, virtual patients for collective
uses of health data, and individual predictions to support clinical decisions.
2.8 Digital Twins 13
•Multimodal dashboards serve primarily as perceptual and documentation tools for
healthcare providers. By integrating various data sources, including electronic health
records, imaging, and telehealth data, these dashboards offer clinicians a comprehensive
view of the patient’s status. However, Demuth et al. highlight that such tools are inher-
ently retrospective and constrained by regulatory considerations related to data privacy.
•Virtual patients are synthetic representations generated from real patient data using
statistical models or machine learning techniques. These are designed for secondary uses
such as research, training, or algorithm development, offering an anonymized alternative
that avoids the limitations of direct data sharing. However, their generation involves a
critical trade-off between privacy and utility, where synthetic data must be sufficiently
destructive to protect patient privacy while maintaining sufficient similarity to real data
for its intended use.
•Individual predictions constitute the prospective element of the framework, where per-
sonalized forecasts derived from predictive analytics support clinical decision-making.
These predictions are context-specific and depend heavily on appropriate preprocessing
and modeling of patient data tailored to specific clinical scenarios.
The authors emphasize that for a given patient, multiple digital representations may be
generated according to the different clinical pathways the patient encounters, with each repre-
sentation tailored to balance the trade-offs associated with specific points of care and intended
uses. The framework positions clinical pathways as central organizing structures that deter-
mine which type of digital representation is most appropriate for each clinical context Demuth
et al. (2025).
Integration Levels
A foundational framework proposed by van der Valk et al. proposes key distinctions between
the levels of integration between the physical entity and the digital twin van der Valk et al.
(2022). They define three primary archetypes: the digital model, the digital shadow, and the
digital twin. These concepts are differentiated based on the directionality and automation of
data flow between the physical and digital representations.
•Digital Model: A static digital representation of a physical object. There is no automatic
data exchange. Any updates to the model or the physical object must be carried out
manually. This archetype is often used for simulations or design prototypes where real-
time data is not required.
•Digital Shadow: A system in which data flows automatically from the physical object
to the digital representation. Changes in the physical system are reflected in the digital
counterpart, but not vice versa. This allows for real-time monitoring, but does not enable
control or interaction with the physical system.
14 Background
•Digital Twin: A fully integrated system where data flows bidirectionally between the
physical and digital entities. This allows for real-time updates, predictive modeling, and
feedback mechanisms. Changes in one representation can influence the other, creating a
dynamic, interactive loop.
Understanding these levels of integration is crucial when designing and evaluating digital
twin systems. Many applications labeled as "digital twins" in practice may more accurately
be categorized as digital shadows or even digital models, due to the absence of automated or
bidirectional data exchange.
Lifestyle-Focused Archetypes
The framework developed by van der Valk et al. offers a general classification of digital rep-
resentations, including digital models, shadows, and twins van der Valk et al. (2022). It is
designed for broad applicability across various domains, including manufacturing and engi-
neering. While relevant in concept to digital health, this framework does not focus specifically
on clinical or lifestyle contexts. In contrast, the model proposed by Borkenhagen and Babic is
tailored to applications in cardiovascular care and lifestyle-related health interventions Borken-
hagen and Babic (2025).
To better integrate clinical data with long-term behavioral factors, Borkenhagen and Babic
introduce a three-tiered framework of lifestyle-focused digital twin archetypes. The model
aims to enhance existing digital twin systems by incorporating personal lifestyle factors, in-
cluding diet, physical activity, stress levels, and sleep patterns, into simulation and decision-
making processes. This approach emphasizes the importance of behavior in the prevention
and management of heart disease and supports more personalized and preventive healthcare.
The three archetypes are defined as follows:
•Basic Archetype: A foundational model based on self-reported lifestyle indicators, in-
cluding body mass index, dietary habits, and exercise routines. This level is suitable
for individuals seeking general lifestyle guidance and awareness, and requires minimal
technological input.
•Intermediate Archetype: A more advanced configuration that includes real-time data
from wearable devices, such as heart rate monitors and sleep trackers. This model en-
ables dynamic feedback and continuous lifestyle monitoring, intended for individuals
with low to moderate cardiovascular risk.
•Advanced Archetype: A comprehensive model that combines detailed clinical data,
including electronic health records, laboratory results, and diagnostic imaging, with
lifestyle information. This archetype enables predictive modeling, simulation, and indi-
vidualized treatment planning, particularly for high-risk or chronically ill patients.
2.9 Personalized Medicine 15
This structured approach parallels the progression described in the hierarchy by van der
Valk et al. Both models describe an evolution from descriptive systems to more dynamic and
predictive capabilities. However, van der Valk’s framework is rooted in systems engineering,
focusing on data flows and technical integration, while the Borkenhagen and Babic model
applies similar principles in a healthcare context that prioritizes lifestyle, behavior, and patient
engagement.
The framework also highlights several enabling factors and challenges. These include the
integration of behavioral science, the need for collaboration across disciplines, and the tech-
nical difficulty of combining subjective lifestyle inputs with structured clinical data. Despite
these challenges, lifestyle-oriented digital twins represent a valuable step toward more holistic,
personalized, and participatory approaches to digital health.
2.9 Personalized Medicine
The foundation of personalized medicine lies in recognizing that each patient presents a dis-
tinct biological and environmental profile that influences their disease manifestation and treat-
ment response Goetz and Schork (2018). Data-driven healthcare and digital twins provide
enhanced potential for personalized healthcare. Cellina et al. argue that the use of digital
twin technology may significantly advance patient care by supporting a more individualized,
data-driven approach to medical treatment Cellina et al. (2023).
2.10 Public Concern About Climate Change in Norway
Although this master’s thesis primarily focuses on the visualization of health-related content,
Case Study 2 explores how health can be used as a framing element for another critical is-
sue: climate change. More specifically, the study investigates how health status, demograph-
ics, and personal attitudes may influence individuals’ levels of concern about climate change.
These domains were chosen based on the hypothesis that they could play a determining role in
shaping climate-related worry, offering insight into how different population groups perceive
climate risks.
Demographics
Within the demographic domain, the factors considered were age, gender, level of urbaniza-
tion, and county of residence.
Previous research by Hickman et al. on the influence of age on climate change worry
found that younger people often express concern due to a perceived lack of urgency from the
adult generation. Their emotional responses are linked not only to the climate crisis itself
but also to feeling dismissed or ignored by older generations. At the same time, youth have
increasingly taken prominent roles in climate activism, highlighting a complex relationship
16 Background
between generational concern and emotional engagement with the issue Hickman (2020).
Regarding gender, data from the Poushter et al. show that 78% of women in Sweden
consider climate change a major concern, compared to 62% of men Poushter et al. (2022).
However, this gender gap is not consistent worldwide. Bush and Clayton observed that gender
differences in climate concern are primarily present in wealthier countries. Their analysis
found a strong correlation between a country’s GDP per capita and the size of the gender gap:
in high-income countries, women tend to express more concern about climate change than
men, while in lower-income nations, the difference is much smaller or nonexistent Bush and
Clayton (2022).
In terms of urbanization and the county, the Nordregio Report reveals a clear difference
in climate concern between urban and rural populations across the Nordic region. According
to the survey underlying the report, 82% of urban residents agreed that climate change is a
serious or very serious problem, compared to just 62% of rural residents Tapia et al. (2023).
Health
In the health domain, the selected variables were satisfaction with physical health, long-term
illness or heart conditions, and rating of overall health. Previous research by Chain, Chain and
Pelliccia fuound that individuals with chronic illnesses are generally more likely to express
heightened concern regarding climate change, particularly due to its perceived impacts on
public health and vulnerability Chain et al. (2022).
Attitude
The attitude category analyzed variables such as determination, life control, and problem solv-
ing. Previous research has shown that self-efficacy, along with knowledge and attitudes, is a
strong predictor of climate risk perception Hidalgo et al. (2010).
Chapter 3
Methodology
This chapter presents the research methodology applied across the different case studies. It
outlines the theoretical foundation that informed the development and evaluation of the arti-
facts created. The thesis adopts design science as its primary research paradigm, which em-
phasizes the creation of innovative solutions to real-world problems through iterative design
and evaluation.
The following sections describe the key concepts of design science, including iterations,
practices, and guidelines. The chapter also introduces the case studies and the analytical stud-
ies conducted in case studies 2 and 3.
Design principles specific to infographics and dashboards are presented. Additionally, a
discussion of prototyping as a development tool and the importance of user-centered design in
guiding prototype development.
Furthermore, the chapter describes the data collection methods used, including focus
groups and surveys, and introduces the concept of multi-audience communication, highlight-
ing its significance in the design process. Finally, the chapter addresses the collaborative efforts
that contributed to the study.
3.1 Design Science
Design science is a research approach that focuses on developing and creating artifacts to
solve practical problems in information systems (Hevner et al., 2004). These artifacts, which
can be objects, methods, or designs are tailored to specific challenges. They can be physical,
conceptual, or digital. In IT, artifacts range from formal systems to everyday applications,
such as social media and gaming tools (Johannesson and Perjons, 2021). By developing and
introducing new artifacts, design science drives change. However, this focus also presents a
limitation, as it is primarily constrained by the needs and wants of stakeholders (Johannesson
and Perjons, 2021).
Baskerville suggests that defining what design science is not is a better way to clarify its
boundaries than attempting to find a universal definition (Baskerville, 2008). He lists several
18 Methodology
things that design science is not, including: design, design theory, IT artifacts, methodology,
action research, computer science, a separate academic discipline, or something entirely new
(Baskerville, 2008).
3.1.1 Iterations
A key characteristic of design science is its iterative problem-solving approach, where artifacts
are continuously refined through cycles of development and evaluation (Hevner et al., 2004).
These products are tested in real-world environments to ensure their utility, with each iteration
improving their effectiveness. Because design is iterative and incremental, evaluation provides
essential feedback to the construction phase, ensuring that both the process and artifact meet
required constraints (Hevner et al., 2004). A design is only considered complete when it
successfully addresses the problem it was meant to solve. However, in complex environments
like manufacturing, iterative development introduces challenges. While improvements can be
observed, formal evaluation is difficult since each iteration induces uncontrolled organizational
changes (Hevner et al., 2004). Despite this, ongoing refinement remains essential for adapting
to evolving needs and ensuring robust solutions (Hevner et al., 2004).
3.1.2 Practices
In “An Introduction to Design Science” Johannesson and Perjons emphasize the relationship
between people, practices, problems, and artifacts. They highlight that when people engage in
practices, they inevitably encounter problems, which can often be addressed through artifacts
(Johannesson and Perjons, 2021). Figure 3.1, inspired by Johannesson and Perjons’s original
illustration of the relationship between people, practices, problems, and artifacts, represents
this process (Johannesson and Perjons, 2021). The illustration was created in Canva and
demonstrates how a person, within a practice, perceives a problem and uses an artifact (in this
case, an ERP system) to address it.
3.1.3 Guidelines
Hevner et al. created seven guidelines with the purpose of assisting researchers, reviewers, ed-
itors, and readers to understand the requirements for effective design-science research (Hevner
et al., 2004). Each guideline should be addressed for design-science research to be complete,
and these are summarized in Table 3.1
3.1 Design Science 19
Figure 3.1: Illustration of the relationship between people, practices, problems, and artifacts.
Table 3.1: The seven guidelines for rigorous design science developed by (Hevner et al., 2004).
No.Guidelines Description
1 Design as an Artifact
Design science research must produce a workable,
practical artifact in the form of a construct, model,
method, or instantiation
2 Problem Relevance
The objective of design-science research is to de-
velop technology-based solutions to important and
relevant business problems.
3 Design Evaluation
The utility, quality, and efficacy of a design artifact
must be rigorously demonstrated via well-executed
evaluation methods.
4 Research Contributions
Effective design-science research must provide
clear and verifiable contributions in the areas of the
design artifact, design foundations, and/or design
methodologies.
5 Research Rigor
Design-science research relies upon the applica-
tion of rigorous methods in both the construction
and evaluation of the design artifact.
6 Design as a Search Process
The search for an effective artifact requires utilizing
available means to reach desired ends while satis-
fying laws in the problem environment.
7 Communication of Research
Design-science research must be presented ef-
fectively both to technology-oriented as well as
management-oriented audiences.
20 Methodology
3.2 Threats to validity
Measures were taken to ensure research validity and data integrity:
•Ethical Approval: Ethical clearance was obtained for handling participant data in the
survey and using a dataset from their survey bank in Case Study 2.
•Data Contracts: A formal agreement was signed between the University of Bergen
and the All of Us database for the use of data in Case Study 3. Personal certifications
to access the data were obtained by completing the courses Responsible Conduct of
Research and Researcher Workbench: Controlled Tier Data.
3.3 Case Study
Three case studies were conducted focusing on different themes, with similar visualization
forms.
•Case Study 1: Dietary Guidelines
Focused on developing infographics to communicate Norwegian national dietary guide-
lines. The case included participant involvement (n=72), and feedback was collected
through questionnaires and semi-structured interviews.
•Case Study 2: Climate Change Concerns
Visualizations, including infographics and interactive dashboards, were created using
data from the Quality of Life Survey (2023). Participants at the SDG Conference 2025 in
Bergen were recruited to evaluate the designs and provide insights into the use of digital
twins.
•Case Study 3: Digital Twins in Healthcare
This case involved creating digital twin prototypes based on cluster analysis of health
data from the All of Us database. Unlike the previous cases, this study did not include
user participation, but rather evaluated how the work fit into existing digital twin frame-
works.
3.4 Analytical study
The analytical foundation of this study lay in the data analysis conducted for Case Studies 2
and 3.
In Case Study 2, data from the Quality of Life Survey (2023) were analyzed using SPSS to
conduct an exploratory analysis. The results from the analysis were used for infographics and
dashboard development.
3.5 Design Principles 21
In Case Study 3, Borkenhagen analyzed data from the All of Us Database, creating popu-
lation clusters based on health status. Based on that, visualizations of the clusters were used
to develop digital twin prototypes.
3.5 Design Principles
Design science research seeks both to develop artifacts and generate knowledge about them
Johannesson and Perjons (2021). This includes design principles, which are prescriptive in-
sights derived from artifacts that can guide future designs across various contexts Johannesson
and Perjons (2021). Design principles serve to guide designers on what elements to include or
exclude in the interface, helping them to consider various aspects of their designs Sharp et al.
(2019).
3.5.1 Design Principles for Infographics
The seven GRAPHIC Principles of Public Health Infographic Design are evidence-based
guidelines developed by Catherine Stones and Mike Gent to support the creation and com-
missioning of health infographics for public audiences Stones and Gent (2014). They include:
•G – Get to know your Audience
•R – Restrict Color
•A – Align Elements
•P – Prioritize Parts
•H – Highlight the Heading
•I – Invest in Imagery (wisely)
•C – Choose Charts Carefully
3.5.2 Design Principles for Dashboards
In Dashboard Design Patterns, Bach et al. gather design principles particular to dashboards
that are commonly agreed upon by several case studies and scholars Bach et al. (2022). They
found a collection of guidelines that emphasize what effective dashboards should and should
not do.
Dashboards should:
•Carefully select key performance indicators (KPIs).
•Align with existing workflows.
22 Methodology
•Include both functional features and visual features.
•Provide consistency.
•Provide interaction affordances.
•Manage complexity.
•Organize charts symmetrically.
•Group charts by attribute.
•Clearly separate groups of charts.
•Order charts chronologically.
Dashboards should avoid:
•Overwhelming users.
•Visual clutter.
•Poor visual design.
•Show too much data.
3.6 Prototyping
Prototypes are tools that help confirm or reject ideas by making them concrete and testable.
Users often struggle to articulate what they want, but when given the chance to interact with
a prototype, they can quickly identify what doesn’t work and more easily provide feedback
through hands-on exploration Johannesson and Perjons (2021); Sharp et al. (2019). The way
we use prototypes during the design process can lead to a variety of new insights and even
inspire new ways of thinking Auflem (2023). These ideas often go beyond what the prototype
itself shows, as they are based on our assumptions and the reasoning we apply Auflem (2023).
3.7 User-Centered Design
In Interaction Design: beyond human-computer interaction Sharp, Rogers and Preece argue
that a user-centered approach is more a philosophy than a technique. User-centered design
places the needs, goals, and experiences of real users at the heart of product development,
rather than letting technology alone dictate the outcome. A well-designed system should en-
hance human skills and judgment, support users in their tasks, and fit naturally into the context
of use, rather than restricting or complicating it Sharp et al. (2019). The concept of user-
centered design is based on three core principles: focusing early on users and tasks, relying on
empirical measurement, and embracing an iterative design process Sharp et al. (2019).
3.8 Data Collection 23
3.8 Data Collection
In Case Study 1 6, feedback gathered from users was used to improve the infographic pro-
totypes iteratively. In Case Study 2 7, a survey was conducted during the SDG Conference
workshop 10.4. This survey was conducted to gather insights into health-related visualiza-
tions and public’s opinion on the use of digital twins. These responses laid the groundwork for
Case Study 3 8. Additionally, the data collected across case studies provided valuable support
for addressing the overall research questions by capturing user attitudes, comprehension, and
engagement.
3.8.1 Focus Group
In Case Study 1, focus groups were used as a qualitative method to gather insights from par-
ticipants representative of the target population. A typical focus group involves between three
and ten individuals and is facilitated by a trained moderator who guides the discussion based
on a preset agenda Sharp et al. (2019). The moderator plays a crucial role in steering the
conversation, encouraging quieter participants to contribute, and managing more dominant in-
dividuals to ensure balanced participation Sharp et al. (2019). This method is particularly
effective for uncovering shared concerns, exploring diverse perspectives, and identifying po-
tential conflicts or misunderstandings Sharp et al. (2019). By fostering open dialogue, focus
groups can surface issues that might be missed in individual interviews Sharp et al. (2019).
3.8.2 Survey
A survey is a systematic method of data collection used to gather information from a prede-
fined group of respondents. They can include both closed and open-ended questions and are
especially effective for reaching a large number of participants with minimal resources Sharp
et al. (2019). This makes them suitable for collecting broad insights, especially when partici-
pants are geographically dispersed or have limited availability. Surveys require careful design
to ensure clarity and the possibility of analyzing the responses. Well-crafted surveys help
gather specific data efficiently from a large group of users Sharp et al. (2019).
3.9 Multi-Audience Communication
The designs presented in the case studies embrace a multi-audience approach, acknowledg-
ing that scientific and medical information must be accessible to users with varying levels of
expertise. Designing for multiple audiences involves more than simplifying content. It re-
quires deliberate choices in structure, narrative, and visual strategy to ensure comprehension,
engagement, and relevance across user groups Mörth (2022).
24 Methodology
This approach reflects a broader shift in visualization research that emphasizes presentation
as a key component of scientific work. Designing with multiple audiences in mind ensures that
both experts and non-experts can meaningfully engage with the content, enhancing the reach
and impact of the communication Mörth (2022).
3.10 Collaboration
Although this master’s project has been written independently, parts of the work, particularly
those related to case study 3, have been developed in close collaboration with Anders Borken-
hagen Borkenhagen (2025). Our shared focus on digital twins facilitated an interdisciplinary
exchange, where Borkenhagen has contributed by developing code for machine learning, PCA
scores, and cluster analyses. I focused on visualizing the outcomes of his analyses to make
them easier to interpret and communicate. Throughout the process, we held frequent meet-
ings, shared relevant academic literature, and exchanged insights on digital twin technologies.
This collaborative process helped strengthen both the methodological and practical aspects of
the project.
Chapter 4
Technologies and Tools
This chapter outlines the main tools and platforms used throughout the project. Each served
a distinct role, whether in developing visual materials, analyzing data, or gathering user feed-
back. The combination of creative and analytical technologies enabled efficient work across
various aspects of this thesis.
4.1 Canva Pro
Throughout this project, Canva Pro has been extensively used as the primary tool for develop-
ing infographics and dashboards in the different case studies. Launched in 2013, Canva is an
Australian platform designed to make visual communication accessible (Canva, 2024). With
its intuitive drag-and-drop interface and rich library of templates, fonts, images, and other cre-
ative assets, it enables users, whether professionals or those without formal design training,
to bring ideas to life from any device (Canva, 2025). Canva simplifies the creative process,
allowing researchers in this project to rapidly develop professional-quality visuals and focus
more on content rather than technical design skills (Perez, 2013). Using Canva made it easy to
produce professional-quality visuals rapidly.
4.2 Flourish
Flourish has been used to add interactive elements to dashboards, particularly in creating in-
teractive visualizations such as the map of Norway shown in 7.17. It is well-suited for making
visualization stories and interactive features. Flourish was launched in 2018 and became part
of Canva’s company group in 2022, which means that it is very easy to implement designs
made in Flourish in Canva (Clark, 2022; Flourish, 2025).
26 Technologies and Tools
4.3 SPSS
IBM SPSS Statistics was used for exploratory data analysis in Case Study 2, including the
application of one-way ANOVA, meaning analysis of variance, to examine differences in cli-
mate change worry across various factors. SPSS (Statistical Package for the Social Sciences)
is part of IBM’s software platform, offering advanced statistical analysis along with machine
learning algorithms, text analysis, and integration with big data. Its ease of use, flexibility, and
scalability make it accessible to users of different skill levels, which makes it an excellent fit
for this project (IBM, 2025).
4.4 Chatgpt
The free version of ChatGPT (4o) OpenAI (2025) was used during the thesis writing as a
supplementary tool to improve sentence structure. Additionally, ChatGPT (4o) was used to
assist with LaTeX formatting, particularly for creating tables in Overleaf. It was not used to
generate original content, but rather to refine language and provide alternative phrasings.
4.5 SurveyXact
As part of a workshop held at the SDG conference in Bergen in 2025, a survey was conducted
to gather participants’ thoughts on health-related visualization and the use of medical digital
twins 7. For this purpose, SurveyXact SurveyXact (2025) was used to design, distribute, and
analyze the survey.
Chapter 5
Datasets
This chapter presents the datasets that provide the data exploration in Case Studies 2 and 3.
These datasets were selected based on their relevance to health and their potential to support
meaningful analysis and visualization. The following sections offer an overview of the Quality
of Life Survey and the All of Us Research Program, describing their scope, structure, and
specific application within this project.
5.1 Quality of Life Survey (2023)
For Case Study 2 (7), the Quality of Life Survey 2023 Statistisk sentralbyrå et al. (2024) was
used to analyze the relationship between climate concern, demographics, attitudes, and health.
This survey, conducted by Statistics Norway Statistisk sentralbyrå (2025) on behalf of the
Norwegian Directorate of Health Norwegian Directorate of Health (2025a), was obtained from
Sikt’s Survey Bank Sikt (2025). The Survey Bank is a comprehensive archive of research data
from the social sciences, humanities, medical, and health research, where users can explore a
vast collection of survey questions and variables from studies conducted as far back as 1957
Sikt (2025).
The Quality of Life Survey is an annual online questionnaire that collects information on
subjective well-being and living conditions in Norway. Each year, a randomly selected sample
of 40,000 individuals is invited to participate. In 2023, 45% of the selected responded Sikt
(2023).
In addition to the questionnaire responses, various registry data are linked to the Quality
of Life Survey, including information from the National Population Register, education and
income registers, NAV social benefits and employment records, as well as occupation and
salary data Sikt (2023).
The Quality of Life Survey is a well-established survey that has been widely used in Nor-
wegian research. It has informed studies on various topics, including work, health, and societal
adaptation during and after the COVID-19 pandemic Ugreninov et al. (2023), loneliness and
social isolation Barstad (2021), and the demographics most affected by negative life events
28 Datasets
Barstad (2022).
5.2 All of Us
The All of Us Research Program is an initiative that aims to collect and analyze health data
from over one million people in the United States All of Us Research Program (2021). All of
Us seeks to accelerate health research and medical advancements. By collecting various health
data, the program supports the development of individualized prevention strategies, treatments,
and care plans tailored to different populations All of Us Research Program (2021).
The program operates by gathering participant data from various sources, including sur-
veys, electronic health records, and biosamples. This data is transferred to a cloud-based
environment, where it undergoes a structured curation process within a protected data reposi-
tory. From there, the curated data is made accessible through the Research Hub, which serves
as the central platform for exploring available datasets. Although anyone can browse the Re-
search Hub, only registered researchers are permitted to access the Researcher Workbench,
provided they verify their identity and complete the program’s access procedures. Access to
the data also requires completing extensive training, including two mandatory courses: Re-
sponsible Conduct of Research and Researcher Workbench: Controlled Tier Data. In addition,
institutions must undergo an administrative process to enable their employees and students to
access. Only researchers affiliated with institutions that have signed a Data Use and Regis-
tration Agreement (DURA) are eligible to register for the All of Us Researcher Workbench.
Within this secure platform, researchers can build cohorts, create collaborative projects, and
use interactive analysis tools All of Us Research Program (2025a).
The All of Us dataset was chosen specifically for its large amount of data from wearables,
which makes it ideal for experimentation with digital twin visualizations. As of 2025, it in-
cludes Fitbit data from nearly 60,000 people, making it the largest publicly available dataset of
its kind. Some of the records even date back more than ten years, which is very useful for ex-
amining long-term trends All of Us Research Program (2025b). The All of Us dataset has been
widely used in medical research. To assess its impact, a search for “All of Us Research Pro-
gram” in Google Scholar, PubMed and ScienceDirect was conducted, with results limited to
the period from 2018 to 2025, as the program began in 2018. This search returned 7063 results
on Google Scholar, 734 on PubMed, and 737 on ScienceDirect, highlighting the extensive use
of the dataset in recent research. The evolution of the number of publications containing “All
of Us Research Program” can be seen in Figures 5.1, 5.2, and 5.3.
5.2 All of Us 29
Figure 5.1: PubMed.
30 Datasets
Figure 5.2: Google Scholar.
5.2 All of Us 31
Figure 5.3: Science Direct.
32 Datasets
Chapter 6
Case Study 1: Enhancing Public Engage-
ment with Dietary Guidelines Through In-
fographics
6.1 Introduction
Understanding healthy nutrition advice is essential for promoting public health, but dietary
guidelines are often presented in dense, text-heavy formats that can limit comprehension and
reduce public engagement. This case study investigates how infographics, as a visual commu-
nication strategy, can improve the accessibility and impact of the Norwegian National Dietary
Guidelines.
A series of infographic prototypes was created based on the guidelines issued by the Nor-
wegian Directorate of Health. The study aims to determine whether visual formats could
enhance understanding and increase engagement with nutritional recommendations. It also
considers how different visual styles might appeal to various demographic groups, recogniz-
ing that a single design is unlikely to be equally effective for all audiences. Therefore, the
project includes several infographic variations with different layouts and visual emphases to
evaluate their potential across a diverse population.
This work led to the paper Tailoring Infographics of the Norwegian National Dietary
Guidelines, which was presented at the ICIMTH 2024 Conference Andreassen and Babic
(2025a).
6.2 Background
The Norwegian Directorate of Health, operating under the Ministry of Health and Care Ser-
vices, provides national dietary guidelines to promote public health Helsedirektoratet (2017).
In August 2024, the Directorate released updated nutrition recommendations aimed at the
general population. However, due to resource constraints, outreach efforts primarily focus on
34
Case Study 1: Enhancing Public Engagement with Dietary Guidelines Through
Infographics
Figure 6.1: A customized version of Figure 3.1 created to illustrate the process of identifying a problem
and addressing it through the development of an artifact, in this case infographics.
young people, while older populations rely on text-based materials, which may limit accessi-
bility and retention NTB (2024).
Effectively communicating dietary information remains essential, given nutrition’s role
in the prevention of chronic diseases World Health Organization and Food and Agriculture
Organization of the United Nations (2003).
6.3 First Iteration
To explore the impact of infographic design on the communication of dietary guidelines, the
same nutritional information was presented using three distinct infographic prototypes, cre-
ated in Canva Pro Andreassen and Babic (2025a). The first iteration of the case study involved
developing these prototypes and facilitating focus group evaluations to assess their effective-
ness.
Infographic 1 emphasizes simplicity and clarity. It employs basic food illustrations and text
organized into structured boxes. This design aims to present the dietary guidelines straightfor-
wardly, ideal for readers seeking quick comprehension. Its simplicity may cater to those who
value direct and uncomplicated information presentation Andreassen and Babic (2025a).
6.3 First Iteration 35
Figure 6.2: Infographic 1.
Infographic 2 adopts a more visually dynamic approach, utilizing a colorful, segmented
wheel layout. This prototype does not incorporate food illustrations, but rather uses its seg-
ments to organize the text in a circular pattern, designed to resemble a food plate. The design
seeks to engage readers visually while offering a structured representation of the dietary guide-
lines. It aims to appeal to those who appreciate an organized yet visually stimulating format,
though the unconventional layout may present a challenge to readers accustomed to a linear
format Andreassen and Babic (2025a).
36
Case Study 1: Enhancing Public Engagement with Dietary Guidelines Through
Infographics
Figure 6.3: Infographic 2.
Infographic 3 focuses on aesthetics, using detailed food illustrations to create an immersive
visual experience. This design eliminates text boxes, placing the text directly beneath the
illustrations to create a more open and visually integrated layout. It is tailored for readers who
prioritize visual engagement Andreassen and Babic (2025a).
6.3 First Iteration 37
Figure 6.4: Infographic 3.
6.3.1 Prototype evaluation
The evaluation of the prototypes was conducted through focus group sessions, stratified by
age groups: 10-29, 30-49, 50-69, and 70-89 years. A total of 72 participants were recruited
through personal contacts and email invitations, with the majority being affiliated with the
University of Bergen Andreassen and Babic (2025a).
Each focus group session involved participants reviewing the three infographic prototypes
alongside the original text format of the dietary guidelines, as presented on the Directorate’s
website Helsedirektoratet (2017). The sessions were facilitated with structured guidance to
support the discussions and to collect both verbal and written feedback. Detailed notes were
taken to document participants’ impressions, preferences, and suggestions regarding the ef-
fectiveness of each infographic design in conveying the dietary guidelines. The prototypes
38
Case Study 1: Enhancing Public Engagement with Dietary Guidelines Through
Infographics
were produced in both Norwegian and English, and participants were shown the version that
matched their preferred language Andreassen and Babic (2025a).
To present the quantitative results, Figure 6.5 was used Andreassen and Babic (2025a). The
figure features multiple person-shaped charts with percentage fills, illustrating the distribution
of preferences across the three infographic prototypes. This visualization helps show which
infographics were most popular among different age groups. Three distinct colors were used
to represent the different infographics, making it easier to compare proportions. Some figures
are filled with all three colors, indicating a high level of disagreement within those groups.
The qualitative results, including descriptions of each age group’s preferences and selected
direct quotes, were organized in Table 6.1 to present and compare feedback across the different
age groups clearly.
Figure 6.5: Focus group voting distribution Andreassen and Babic (2025a).
6.3 First Iteration 39
Table 6.1: Focus group participant qualitative feedback on each infographic by age group Andreassen
and Babic (2025a).
Age Group Infographic 1 Infographic 2 Infographic 3
10–29
(n=40)
Highlighted for its
“great balance be-
tween illustrations
and text,” which
promoted engage-
ment. However,
one participant
noted it felt “too
conventional and
boxed-in.”
Not favored as a top
choice. Participants
mentioned missing illus-
trations and disliked the
text background color.
Positive feedback in-
cluded its ability to “divide
and categorize guide-
lines” and being “the most
scientific and easy to un-
derstand.”
Appreciated for “appealing il-
lustrations” and being “engag-
ing without much reading.”
Criticized for lacking struc-
tured content. Many found it
the most visually pleasing in-
fographic, but desired better
organization.
30–49
(n=16)
Praised for being
the “best struc-
tured,” with a
“good combina-
tion of text and
data” and a “nice
overview.” De-
scribed as “the
most ordered” and
visually appealing.
Mixed feedback, with
one participant liking "the
structure and clear mes-
saging", but feeling that
it lacked illustrations.
Suggested adding food
images, possibly in the
background.
Praised for its “fresh mes-
sage” and visual appeal. One
participant described it as “or-
dered, but not too much,”
while another highlighted its
“fine combination of text and
illustrations.” Some found it
“a bit busy, but not overly dis-
organized.” One participant
likened it to a “mind map,” ap-
preciating how it “best com-
bines illustrations and infor-
mation.”
50–69
(n=12)
Noted for “seam-
lessly combining
text and visuals,”
making it easy to
understand. One
noted that “the
color choice of
green background
fits the theme of
eating healthy.”
Seen by some as having
a “professional feel.” Par-
ticipants mentioned it “fa-
cilitated categorization of
dietary advice” and “pro-
vided a sense of compe-
tence and authority.” How-
ever, several mentioned
that they missed illustra-
tions.
Reactions were mixed.
Some felt it needed no
changes, while one called it
“old-fashioned.” Criticisms in-
cluded the “illustrations in the
corner” and “food images of
milk in glass bottles, donuts,
and macaroons as sweets
do not fit Norwegian associa-
tions.”
70–89
(n=4)
Viewed as “well-
organized and
easy to follow,”
aiding memory re-
tention of dietary
guidelines.
Not favored due to be-
ing difficult to read, with
one participant noting
that “this was the only
infographic they needed
glasses to read.”
Liked for the illustrations, but
criticized for the lack of order.
One said, The food in the il-
lustrations looks so good, I
was motivated to eat healthier
right away!”
40
Case Study 1: Enhancing Public Engagement with Dietary Guidelines Through
Infographics
Notably, although Infographic 3 was often praised for its visual appeal, it was also crit-
icized for lacking structural clarity, a contrast that highlights the need to balance aesthetic
appeal with clear communication.
6.4 Second Iteration
In the second iteration of the case study, feedback from the focus group sessions was analyzed
and used to refine two of the original infographic prototypes.
Infographic 1 remained unchanged, as participants responded positively to its structure and
clarity, and no substantial revisions were deemed necessary.
Infographic 2 was revised to better meet user preferences by incorporating additional illus-
trations, using a lighter and more visually appealing color palette, and removing text shadow
behind to improve readability. These changes directly addressed feedback highlighting a de-
sire for more engaging visuals and a clearer layout.
6.4 Second Iteration 41
Figure 6.6: Infographic 2 after second iteration.
Infographic 3 underwent more extensive changes, particularly in terms of content struc-
ture. Inspired by the balance and organization of Infographic 1, the layout was redesigned to
improve flow and comprehension. Cultural relevance was also considered in this iteration; il-
lustrations were updated to reflect Norwegian associations more accurately, such as replacing
generic icons with imagery more closely related to milk and sweets. These refinements aimed
to increase both the cultural resonance and the visual clarity of the infographic Andreassen
and Babic (2025a).
42
Case Study 1: Enhancing Public Engagement with Dietary Guidelines Through
Infographics
Figure 6.7: Infographic 3 after second iteration.
6.4.1 User-Centered Design
This case study employed a user-centered design approach, prioritizing the needs, prefer-
ences, and feedback of the intended audience in the development of visual communication
artifacts. Recognizing that a single design might not resonate equally across diverse demo-
graphic groups, the project engaged end users through age-stratified focus groups. This en-
sured that the design process included infographics that varied in style and layout to match
different audience preferences.
The infographic prototypes aimed not only to convey nutritional information but also to
promote healthier eating behaviors. By using visually engaging formats, they make the infor-
mation more accessible and appealing.
This approach aligned with the three core principles of user-centered design as outlined by
6.4 Second Iteration 43
Sharp, Rogers, and Preece Sharp et al. (2019). The process prioritized early user involvement,
gathered empirical feedback through focus groups, and followed an iterative development cy-
cle where designs were refined based on user input.
6.4.2 Prototyping in the Design Process
In this case study, prototyping played a key role in the design development. When participants
engaged directly with the prototypes, it made it easier for them to identify unclear elements,
suggest improvements, and express preferences they might not have articulated otherwise.
Reflecting the idea that users often struggle to verbalize their needs until they can interact
with a concrete design Sharp et al. (2019). Several improvements resulted from the process of
confirming or rejecting ideas through direct testing and evaluation, including more culturally
relevant imagery, clearer layout organization, and adjustments to tone and color.
Beyond these specific changes, prototyping encouraged broader reflection. As Auflem
notes, prototypes can inspire new ways of thinking and reveal insights that extend beyond the
artifact itself Auflem (2023). In this project, user feedback not only guided refinements but also
prompted a reconsideration of how visual communication can more effectively support public
understanding of health information.
6.4.3 Application of GRAPHIC Principles
The infographic development was guided by the GRAPHIC Principles of Public Health Info-
graphic Design Stones and Gent (2014). These seven principles supported decision-making
regarding layout, structure, and use of visual elements.
44
Case Study 1: Enhancing Public Engagement with Dietary Guidelines Through
Infographics
Table 6.2: Application of GRAPHIC Principles in Infographics in Case Study 1.
Principle Application in the Case Study
Get to Know Your Audience User-centered design involved focus groups from vari-
ous age groups, allowing the infographics to be tailored
to a multi-audience. The user feedback informed refine-
ments.
Restrict Color Infographic 1 uses green tones associated with healthy
eating; Infographic 2 employs harmonious yet distinct col-
ors for each segment; Infographic 3 features soft, warm
pastels to convey a friendly tone.
Align Elements Infographic 1 employs a boxed grid for clarity. Infographic
2 arranges text in a plate-like pattern for visual interest.
Based on feedback, Infographic 3 adopted a similar grid
layout to Infographic 1 for improved organization.
Prioritize Parts Equal emphasis is conveyed through uniform text and
illustration sizes across all guidelines within each info-
graphic.
Highlight the Heading Each infographic emphasizes the “Dietary Advice” head-
ing using bold typography and high color contrast, ensur-
ing immediate clarity of purpose.
Invest in Imagery (Wisely) Infographic 1 features basic food illustrations; Infographic
2 initially lacked images but later introduced simple icons
in response to user feedback; Infographic 3 incorporates
hand-drawn-style food imagery.
Choose Charts Carefully Since the content is not data-driven, no charts were in-
cluded.
6.5 Limitations
The study’s reliance on focus group participants, primarily recruited from the University of
Bergen, limit generalizability to the broader Norwegian population. There was also consid-
erable variation in the number of participants across age groups, with a much larger sample
from the 10–29 age group (n = 40) compared to the 70–89 age group (n = 4). This reflects the
relative ease of recruiting students and university staff, while reaching older participants, espe-
cially those outside digital or academic networks, proved more challenging. This limitation is
particularly significant, given that the study’s motivation stemmed from the concern that cur-
rent digital outreach strategies primarily target younger audiences, leaving older generations
underrepresented.
In addition, while the study explored visual strategies for communicating dietary recom-
6.6 Conclusion 45
mendations, it should be noted that the Norwegian Directorate of Health already provides a
range of professionally developed infographics and materials related to national dietary guide-
lines Norwegian Directorate of Health (2025b). Consequently, the development of new visuals
in this study was not intended to replace existing resources, but rather to serve as a means of
testing alternative designs and communication styles. Future research should aim to include
a more balanced representation of age groups and follow participants over time to assess how
well they retain the information and whether it leads to lasting changes in their eating habits.
6.6 Conclusion
This case study demonstrates how visual design can enhance the appeal of health information
by translating guidelines into infographics tailored to various audiences.
Participants shared that visuals combining clear layout, structured text, and simple illustra-
tions helped them understand the advice more easily. This reinforces the idea that public health
messaging, particularly when intended for broad distribution by institutions like the Norwegian
Ministry of Health, should focus on user-friendly design that connects with diverse population
groups.
The goal was not to replace the original guidelines but to support them with a more en-
gaging format. With input from users and attention to clarity, the infographics evolved into
practical tools for communication, rather than merely decorative elements. There is still more
to learn, particularly regarding the long-term effects. Still, this work demonstrates that visu-
alizing health information can help it reach a broader audience and support more informed
decisions regarding nutrition.
46
Case Study 1: Enhancing Public Engagement with Dietary Guidelines Through
Infographics
Chapter 7
Case Study 2: Visualizing Factors Influ-
encing Climate Change Concern
7.1 Introduction
Drawing on data from the 2023 Quality of Life Survey 5.1, this study examined the associ-
ations between concern about climate change and three key domains: health status, demo-
graphic characteristics, and individual attitudes. The analysis was conducted in two stages.
First, a bivariate exploratory data analysis was performed, examining associations between
concern about climate change and individual variables across three domains: health status,
demographic characteristics, and personal attitudes. Second, selected findings were synthe-
sized and presented through infographics and a dashboard. These visualizations formed part
of the broader exploratory process, supporting the identification and communication of key
patterns in the data. The results informed the design of a workshop held at the Sustainable
Development Goals (SDG) Conference 2025 in Bergen Babic et al. (2025).
7.2 Background
The Quality of Life Survey comprises a wide range of questions; therefore, the initial phase of
this project involved carefully reviewing the survey items to identify those relevant to health,
attitudes, and demographics. These domains were selected based on the hypothesis that they
might significantly influence individuals’ levels of climate worry.
Particular emphasis was placed on the question: “To what extent do you worry about
climate change?”. Responses to this variable were analyzed using one-way ANOVA (analysis
of variance) to examine differences in mean worry levels across various demographic, health-
related, and attitudinal categories. In addition, mean-value charts were plotted to represent
these differences, using IBM SPSS Statistics visually IBM (2025). Special focus was given to
respondents who reported being either “very worried” or “not worried at all,” as these groups
provide the most evident contrast in perspectives. The results were presented through a series
48 Case Study 2: Visualizing Factors Influencing Climate Change Concern
of graphical visualizations, highlighting key patterns and relationships between concern about
climate change and the selected variables.
7.3 Demographics
In the demographics category, the selected variables included:
•Age (Figure 7.1)
•Gender (Figure 7.2)
•Level of urbanization (Figure 7.3)
•County (Figure 7.4)
These factors were analyzed to identify demographic patterns associated with concern about
climate change.
The analysis of the age groups revealed a relatively even distribution of climate concern
across most brackets, with both “not worried at all” and “very worried” responses present.
Although respondents aged 90–100 showed a slight increase in worry, the small sample size
limits the reliability of this result.
In terms of gender, males and individuals identifying as other gender identities were more
represented among those “not worried at all,” whereas non-binary respondents were the largest
group in the “very worried” category. Female respondents were the largest group in the “a
little worried” and “fairly worried” groups. This pattern aligns with findings by Poushter et
al., which indicate that women tend to express greater concern about climate change than men
Poushter et al. (2022).
Urbanization also influenced concern levels. Respondents from more urbanized areas,
particularly central Oslo, expressed higher climate worry, while rural respondents were more
often “not worried at all”. This trend matches patterns described in the Nordregio Report Tapia
et al. (2023).
Lastly, regional differences appeared: Oslo respondents reported the highest worry levels,
while those from Rogaland, Agder, and Troms og Finnmark showed comparatively low con-
cern. These regional variations may reflect local cultural, political, or economic influences on
climate perceptions.
7.4 Health 49
Figure 7.1: Age. Figure 7.2: Gender.
Figure 7.3: Level of urbanization. Figure 7.4: Regional distribution.
7.4 Health
In the health category, the variables analyzed were:
•Satisfaction with physical health (Figure 7.5)
•Long-term illnesses or heart conditions (Figure 7.6)
•Rating of overall health (Figure 7.7)
These indicators were used to explore possible connections between health status and levels
of concern about climate change. The results revealed a consistent pattern across all three
indicators: respondents who expressed either very low or very high levels of concern about
climate change often reported poorer health.
For physical health satisfaction, the most significant proportions in both the “not worried
at all” and “very worried” groups were found among those who reported being “not satisfied at
all” with their physical health. A similar trend appeared in the variable assessing overall health,
where the highest shares within both extremes of climate concern were among respondents
who rated their health as “very poor.”
50 Case Study 2: Visualizing Factors Influencing Climate Change Concern
Respondents who reported having long-term illnesses or showed slightly elevated repre-
sentation in both the least and most worried categories, though the differences were less pro-
nounced.
These findings suggest a potential polarization effect, where poorer self-rated physical
health and overall health satisfaction may be associated with both strong concern and substan-
tial disengagement regarding climate change. This partially aligns with the findings by Chain
et al. Chain et al. (2022). However, the present results also indicate a notable proportion of
individuals with chronic conditions who report no concern at all, a pattern not highlighted in
Chain et al.’s study.
Figure 7.5: Physical health satisfaction. Figure 7.6: Long-term illnesses.
Figure 7.7: Overall health rate.
7.5 Attitude
In the attitude category, the analysis included the variables:
•Determination (Figure 7.8)
•Life control (Figure 7.9)
•Problem solving (Figure 7.10)
These variables were selected to examine whether attitudes influence levels of concern
about climate change. A consistent pattern appeared across all three indicators. Individuals
7.5 Attitude 51
who were either not worried at all or very worried about climate change were more likely to
report low levels of personal agency.
In the determination item, those who strongly disagreed with the statement “If I truly set
my mind to it, I can do almost anything” were the most common in both the lowest and highest
concern groups.
A similar trend was found in the life control item. Respondents who strongly agreed with
the statement “I have little control over what happens to me” , were also most frequent among
the least and most concerned individuals.
The same pattern appeared in the problem-solving item. Respondents with low confidence
in their ability to solve problems were more often found in both the lowest and highest levels
of climate concern.
These findings suggest that having low determination, little life control, and low ability to
solve problems may be associated with both very low and very high concern about climate
change. Although Hidalgo et al. demonstrated that such attitudes can be strong predictors of
climate risk perception, the current results reveal a more complex relationship, where these
attitudes are linked to both minimal and increased worry Hidalgo et al. (2010).
Figure 7.8: Determination. Figure 7.9: Life control.
Figure 7.10: Problem solving.
52 Case Study 2: Visualizing Factors Influencing Climate Change Concern
7.6 Key Findings
Regarding the factors, demographics emerged as the most significant influence, particularly
in terms of the level of urbanization. A clear correlation was observed between residing in
non-central areas and a lower concern about climate change. Concerning health and attitude,
individuals who rated their health poorly or exhibited low determination showed higher pro-
portions in the “very worried” and “not worried at all” categories.
7.7 Phrasing Issue
In the Quality of Life Survey 2023, the most common response to the question “To what
extent do you worry about climate change?” was “a little worried”. The limited range of
answer options may have influenced this outcome. With only four choices (“not worried at
all”, “a little worried”, “fairly worried”, and “very worried”), the gap between “not worried at
all” and “a little worried” appears disproportionately large. As a result, respondents who are
only mildly or occasionally concerned may select “a little worried”, since “not worried at all”
might seem too definitive. Adding an option, such as “neither worried nor not worried”, could
allow for a more accurate reflection of varying levels of concern and improve the precision of
the data collected.
7.8 Infographics
To visualize the results of this analysis, three infographics were developed to illustrate key
findings concerning demographics, health, and attitudes. These visualizations aimed to present
the data in a more accessible and engaging way, particularly for use in the workshop setting at
the SDG Conference.
The infographics were designed to strike a balance between clarity and detail. For the
data representations, the same color schemes used in the SPSS bar charts were applied to
maintain visual consistency. A variety of data representations were tested during the design
process, including pie charts, bubble charts, line graphs, histograms, people graphs, and a map
of Norway.
Overall, the infographics effectively conveyed statistical relationships into visually intu-
itive messages, offering viewers immediate insights while preserving the depth and nuance of
the original data.
7.8 Infographics 53
Figure 7.11: Demographics infographic.
54 Case Study 2: Visualizing Factors Influencing Climate Change Concern
Figure 7.12: Health infographic.
7.8 Infographics 55
Figure 7.13: Attitude infographic.
56 Case Study 2: Visualizing Factors Influencing Climate Change Concern
7.8.1 Application of GRAPHIC Principles
The design process drew on the GRAPHIC principles, which outline key considerations for
making information visually effective and accessible Stones and Gent (2014). These principles
shaped choices related to layout, clarity, and the use of graphical elements.
Table 7.1: Application of GRAPHIC Principles in Infographics in Case Study 2.
Principle Application in the Case Study
Get to Know Your Audience The infographics were designed for participants of the
SDG Conference workshop, with a focus on accessibil-
ity for a public audience interested in health and climate.
Language and visuals were adapted to suit a diverse au-
dience.
Restrict Color A consistent color palette was applied across all three in-
fographics. All pie charts, bubble charts, line graphs, his-
tograms, and people graphs use the same color schemes
as their corresponding SPSS bar charts.
Align Elements Each infographic was structured to ensure alignment of
graphical and textual elements.
Prioritize Parts The different categories: demographics, health and atti-
tude were included as text boxes in capitalized letters to
emphasize their importance.
Highlight the Heading Bold, capitalized headings were used in each infographic.
Invest in Imagery (Wisely) Apart from the maps of Norway, no images were in-
cluded, due to the infographics being content-dense and
not wanting to create visual clutter.
Choose Charts Carefully Several chart types were tested during development.
Final selections include pie charts, bubble charts, line
graphs, histograms, and people graphs.
7.9 Dashboards
In addition to the infographics, an interactive demographics dashboard was developed by inte-
grating dynamic elements from Flourish into the design created in Canva Pro.
The design process involved testing several versions of the map of Norway, evolving from
an initial static black map to an interactive version created in Flourish with hover-over ef-
fects and color-coded regions. The color scheme was also refined to improve accessibility,
addressing colors in earlier versions that were not suitable for individuals with color vision de-
ficiencies 7.14, 7.15, 7.16. The dashboard was designed and evaluated by the design principles
proposed by Bach et al., with a focus on ensuring consistency, a symmetrical layout, mean-
7.10 Survey 57
ingful interactivity, and visual clarity Bach et al. (2022). Efforts were made to avoid visual
clutter, which would overwhelm the user and present an excessive amount of data.
Figure 7.14: First prototype. Figure 7.15: Second prototype. Figure 7.16: Third prototype.
Figure 7.17: Fourth prototype.
7.10 Survey
As part of the workshop held at the SDG conference, a survey was conducted using SurveyXact
to gather participants’ views on health-related visualizations and their thoughts on using digital
twins in a medical context, as an intro to the final case study 10.4. The survey was effective in
gathering insights, but the sample size is small (N = 42), and the group of participants in the
18-29 age range is significantly larger than the other groups.
58 Case Study 2: Visualizing Factors Influencing Climate Change Concern
7.11 Use Cases
Figure 7.18 visualizes the design and development process as a clear and structured path illus-
trating how data is formed and utilized. The process begins with the collection of question-
naires and the creation of a well-organized database, both of which are conducted by Statistics
Norway Statistisk sentralbyrå (2025). From this foundation, the data is accessed, analyzed,
and translated into meaningful visual insights. The three dots at the end emphasize that the
applications and opportunities extend beyond those explicitly shown. At the national level,
these findings can be used to assess public needs, inform healthcare strategies, and identify
key factors influencing concern about climate change.
Figure 7.18: The design development process.
7.12 Conclusion
This case study drew on data from the 2023 Quality of Life Survey to examine the relationship
between concern about climate change and demographic characteristics, health status, and in-
dividual attitudes. The analysis suggests that certain patterns could be associated with varying
levels of climate concern. Factors such as level of urbanization, gender identity, self-rated
health, and self-belief appear to influence the responses of individuals. In several instances,
both high and low concern were found to coexist with low self-efficacy or poor health, indicat-
ing that concern about climate change can be complex and influenced by a range of personal
experiences and psychological factors.
To make the findings more accessible, several visual formats were developed. Infographics
effectively communicated key insights, while an interactive demographic dashboard allowed
users to explore the data in greater depth.
The study also raised questions about survey design choices made in the Quality of Life
Survey. The phrasing and limited range of answer options in the climate concern item may
have affected response patterns. Including a more neutral choice, such as “neither worried nor
not worried”, could potentially capture more nuanced attitudes in future surveys.
By combining quantitative analysis with visualization, this project offers insights into how
various factors may be related to climate concern among the public.
Chapter 8
Case Study 3: Designing Static Digital
Twin Prototypes
8.1 Introduction
This case study focuses on developing a static digital twin based on cluster analysis of health
data from the All of Us Research Program. It details the process of data access, pre-processing,
dimensionality reduction, and clustering to identify meaningful health groups, conducted
by Borkenhagen Borkenhagen (2025). The study then focuses on visualizing these clusters
through human body models, creating a multi-modal dashboard and a personalized infographic
prototype. While the model remains static, it offers valuable insights into communicating com-
plex health data in an accessible and user-friendly format, building on the Basic Archetype
level from Babic and Borkenhagen’s Lifestyle-Focused Archetypes Framework.
8.2 Background
This case study uses data from the All of Us Research Program 5.2. To gain access, the
University of Bergen had to initiate an application process, including signing the Data Use
and Registration Agreement (DURA), as it was not initially listed as a registered institution.
A considerable amount of time was spent completing this administrative process. After ap-
proval, the university was granted access to both the Registered Tier and the Controlled Tier
of the dataset. Following this, Borkenhagen and I completed the required courses, Responsi-
ble Conduct of Research 10.4 and Researcher Workbench: Controlled Tier Data 10.4, to gain
permission to work with the full range of All of Us data.
8.3 Collaboration
This case study follows both the methodological approach and the collaborative structure of the
Digital Twin Project conducted by Sahlgaard, Røise, and Sævareid Røise (2024); Sahlgaard
60 Case Study 3: Designing Static Digital Twin Prototypes
(2024); Sævareid (2024). As in that project, the work was divided into distinct components:
the data was processed using machine learning by Borkenhagen, following an approach similar
to that of Sahlgaard. Meanwhile, I was responsible for visualizing the results, a role similar to
that of Sævareid.
The initial phase involved getting familiar with the data and deciding which variables to
extract from the different All of Us datasets. Borkenhagen performed extensive data cleaning
to prepare the datasets for analysis. He then applied Principal Component Analysis (PCA)
to reduce dimensionality and identify underlying patterns by combining correlated variables
into principal components. Based on insights from the PCA, cluster analysis was conducted
to group participants according to shared characteristics. The clustering was performed using
key health-related features identified as fitting the basic archetype digital twin Borkenhagen
and Babic (2025). These included Body Mass Index (BMI), waist-hip ratio, average daily
steps, average daily sleep hours, smoking status, age, diabetes, hypertension, and parental
history of heart attack. This approach helped reveal clusters within the data relevant to the
study objectives. While some of these variables are straightforward, BMI and waist-to-hip
ratio might require further explanation to clarify their significance.
Body Mass Index (BMI) is a commonly used metric for estimating body fat based on an in-
dividual’s height and weight. It categorizes individuals into one of four groups: Underweight
(BMI below 18.5), Healthy (BMI between 18.5 and 24.9), Overweight (BMI between 25.0 and
29.9), and Obesity (BMI of 30.0 or above). While BMI serves as a general indicator of whether
an individual’s weight is within a healthy range, it does not account for important factors such
as muscle mass, bone density, or overall body composition. For this reason, healthcare profes-
sionals typically use BMI as a preliminary measure within a broader assessment framework to
evaluate an individual’s health status (National Heart, Lung, and Blood Institute, 2025).
The waist-to-hip ratio is another measure used to assess body composition. It is calculated
by dividing the circumference of the waist by that of the hips. Higher values may indicate a
greater concentration of abdominal fat, which is associated with an increased risk of cardio-
vascular disease and other health conditions (WebMD, 2023).
To support the selection of the number of clusters, the silhouette score was used as a metric
for evaluating clustering performance. The silhouette score measures how well each data point
fits within its assigned cluster compared to other clusters Scikit-learn. The score ranges from -1
to 1, where values close to +1 indicate well-defined clusters, values near 0 suggest overlapping
clusters, and negative values may indicate misclassified points Scikit-learn.
Clustering was evaluated using values of K ranging from 2 to 6. The corresponding sil-
houette scores are presented in Table 8.1.
8.3 Collaboration 61
Table 8.1: Silhouette scores for different numbers of clusters K.
K Silhouette Score
2 0.2333
3 0.2105
4 0.2335
5 0.2480
6 0.2460
Based on these results, K = 5 was selected for further visualization, as it achieved the
highest silhouette score and therefore indicated the most coherent clustering structure.
Figure 8.1: K=5 Cluster
62 Case Study 3: Designing Static Digital Twin Prototypes
Table 8.2: Cluster characteristics based on mean values of selected health-related features.
Feature Cluster 0 Cluster 1 Cluster 2 Cluster 3 Cluster 4
BMI 33.41 26.84 36.86 26.36 25.45
Waist-hip ratio 0.95 0.89 0.92 0.84 0.79
Avg. daily steps 4803.38 7698.71 5070.52 8032.97 7748.44
Avg. daily sleep hours 5.37 5.73 5.47 5.83 6.00
Smoking status 1.93 0.04 0.07 1.96 0.02
Age 65.40 69.23 51.41 56.73 39.84
Diabetes 0.34 0.13 0.27 0.07 0.04
Hypertension 0.73 0.54 0.54 0.32 0.10
Heart attack in parent 0.28 0.27 0.20 0.20 0.09
Table 8.2 presents the mean values for each cluster across the selected health-related fea-
tures. For the feature smoking status, the values are coded as follows: 0 = does not smoke,
1 = does not currently smoke but has smoked regularly in the past, and 2 = currently smokes
regularly.
8.4 Prototyping
The prototyping process aimed to approximate a static digital twin by visually representing
health clusters through a combination of human body models and icons, presented within a
multi-modal dashboard and a personalized infographic. These visual elements were designed
to convey the results of the cluster analysis in a format that is both accessible and engaging for
a broad audience.
8.4.1 Visualization of Clusters Using Human Figures
The project focused on exploring various methods to visualize clusters in a way that approx-
imates a digital twin. The initial phase involved brainstorming approaches to represent the
different clusters using human body shapes, aiming to distinguish between groups in an intu-
itive and understandable manner. Various human body figures from Canva Pro were tested;
however, it proved challenging to find consistent and comparable figures that effectively rep-
resented the diverse clusters, particularly considering variations in BMI.
To improve the visualization of human figures, the BMI Visualizer developed by the Per-
ceiving Systems Department at the Max Planck Institute for Intelligent Systems was utilized
Max Planck Institute for Intelligent Systems (2013). This interactive tool employs a statisti-
cal model of human body shape derived from thousands of detailed laser scans, generating
realistic male and female body models based on user-inputted height and weight Max Planck
8.4 Prototyping 63
Institute for Intelligent Systems (2013). It allows dynamic visualization of body shape varia-
tions across the BMI spectrum, providing a more accurate and meaningful representation of
the clusters.
Figure 8.2: The different mean BMI’s of the clusters visualized through BMI Visualizer Max Planck
Institute for Intelligent Systems (2013).
8.4.2 Icon Design for Health Features
The next challenge was to represent key health-related features as icons. This was straight-
forward for average daily steps, sleep hours, and smoking status because these behaviors are
associated with universally recognized symbols. However, creating intuitive and straightfor-
ward icon representations for BMI, waist-hip ratio, age, diabetes, hypertension, and parental
history of heart attack was more complex.
The selected icons for these were as follows: for BMI, a bathroom scale with the letters
“BMI” displayed on it; for waist-to-hip ratio, an icon showing a waist and hips wrapped with
measuring tape; for age, a combination of a child icon and an elderly person icon with a
timeline beneath; for diabetes, an illustration of a finger being tested with a glucometer; for
hypertension, a blood drop accompanied by a blood pressure monitor; and for parental history
of heart attack, which was the most challenging, an icon of a grown person holding a child,
overlaid with a heart symbol marked by lines to indicate an heart attack on the grown person.
Since these icons are not entirely self-explanatory, future work should consider adding a hover-
over function that provides brief textual explanations for each icon.
64 Case Study 3: Designing Static Digital Twin Prototypes
Figure 8.3: The chosen icons.
8.4.3 Dashboard Development
Using these elements, the dashboard was developed through extensive experimentation with
various designs and methods for presenting the data and human figures. The goal was to cre-
ate a multi-modal dashboard that provides a quick overview and is easily understandable to
both healthcare professionals and patients, highlighting a multi-audience approach. It was
also important to include features that clearly distinguish the different clusters. The decision
was made to use a color scheme corresponding to each cluster on the human figure to differ-
entiate them visually. These components were then assembled into a multi-modal dashboard.
The development process was informed by Bach et al.’s design principles Bach et al. (2022),
particularly with regard to consistency, clear separation of data for each cluster, and chrono-
logical ordering. Figure 8.5 illustrates the progression from raw clusters to static digital twin
representations.
8.4 Prototyping 65
Figure 8.4: Illustration of the development process from initial clusters to static digital twin visualiza-
tions in the dashboard.
8.4.4 Infographic Development
To demonstrate how the dashboard concept can be applied in a personalized context, a single
infographic was developed based on a representative individual from the dataset. The val-
ues presented in the infographic were modified to ensure the person’s anonymity, while still
reflecting realistic patterns and characteristics.
The infographic provides a visual representation of the individual’s digital twin, based
on their key health-related metrics. It also consists of a description of the cluster to which
they belong, comparisons between their values and the cluster average, and tailored health
recommendations.
Borkenhagen generated the cluster description, cluster comparison, and personalized ad-
vice Borkenhagen (2025).
This approach demonstrates how personalized health information can be effectively com-
municated in a clear and visually engaging format, designed to facilitate a deeper understand-
ing and encourage meaningful lifestyle actions.
8.5 Design Process 67
Table 8.3: Application of GRAPHIC Principles in Infographics in Case Study 3.
GRAPHIC Principle Application in Case Study 3: Static Digital Twin
Get to Know Your Audience The infographic was primarily designed for use in con-
sultations between doctors and patients. Consequently,
medical terminology was deliberately avoided to enhance
accessibility.
Restrict Color Distinct color were assigned to each cluster, and these
were consistently applied to the human figures.
Align Elements Information is organized in a clear top-down structure:
name and ID are placed at the top, followed by the digital
twin figure and symmetrically spaced health icons. Clus-
ter information and comparative values are arranged in
parallel beneath, while personalized advice occupies the
remaining space at the bottom.
Prioritize Parts The digital twin figure and the icons are visually empha-
sized through size. Bolded headings are used to signal
the content of each section and guide the reader’s atten-
tion.
Highlight the Heading The infographic does not fully comply with this principle,
as the heading does not stand out visually from the rest of
the content. Adding a clear title, such as “Digital Twin In-
fographic,” could have improved clarity, but was ultimately
deemed unnecessary for this context.
Invest in Imagery (Wisely) Realistic human body figures generated using the BMI
Visualizer were used to illustrate the patient. Icons were
carefully selected to represent key health variables, help-
ing to reduce textual load while maintaining clarity.
Choose Charts Carefully Instead of traditional charts, visual storytelling was prior-
itized. Human figures and icons served as stand-ins for
quantitative charts.
8.5 Design Process
Figure 8.6 provides an overview of the design process, beginning with cluster identification,
followed by the creation of a multi-modal dashboard, and concluding with the development of
a personalized infographic.
68 Case Study 3: Designing Static Digital Twin Prototypes
Figure 8.6: Overview of the cluster structure, dashboard design, and an example of a personalized
infographic.
8.6 Digital Twin Frameworks
Comparing the developed prototypes with established digital twin frameworks offers valuable
insight into both their conceptual foundations and practical constraints. The following analysis
includes the frameworks introduced in Chapter 2, assessing how well the prototypes align with
each and identifying essential points of deviation.
8.6.1 Data-Centric Framework
Demuth et al. propose a framework that distinguishes between three primary types of digital
representations within medical digital twins: Multi-modal Dashboards, Virtual Patients, and
Individual Predictions. The dashboard prototype presented in this case study partially aligns
with the first category, the multi-modal dashboard. As it does not incorporate dynamic data in-
put, automated updates, or interactive feedback mechanisms, this case may exemplify one of
the scenarios outlined by Demuth et al., in which the term “digital twin” is applied inconsis-
8.7 Limitations 69
tently, as the system does not meet the integration and functional requirements defined in their
framework Demuth et al. (2025).
8.6.2 Integration Levels Framework
According to the Integration Levels Framework proposed by van der Valk et al., digital twin
implementations can be categorized into three archetypes: Digital Model, Digital Shadow, and
Digital Twin. The prototypes in the case study align with the Digital Model archetype, as
they are based on static data without real-time updates. Although it is referred to as a digital
twin, the absence of dynamic integration and feedback functionality means it does not meet
the criteria for the higher archetypes in this framework van der Valk et al. (2022).
8.6.3 Lifestyle-Focused Archetypes Framework
The prototypes presented in this case study were developed with the Basic Archetype of
Borkenhagen and Babic’s Lifestyle-Focused Archetypes Framework in mind, in collabora-
tion with one of the framework’s original authors. This framework defines three archetypes
of lifestyle-oriented digital twins: Basic, Intermediate, and Advanced. The Basic archetype
is characterized by its reliance on self-reported lifestyle indicators. It targets users seeking
general lifestyle guidance and health awareness, and requires minimal technological infras-
tructure. Among the three frameworks considered, this classification is the most accurate and
appropriate for describing the prototypes developed in this case study Borkenhagen and Babic
(2025).
8.7 Limitations
Despite the potential of this case study, several limitations must be acknowledged.
The cluster analysis was based on a selected set of health-related variables, prioritizing
those most relevant to lifestyle and cardiovascular disease risk. The resulting clusters should
therefore be interpreted as simplified representations rather than exhaustive reflections of in-
dividual health profiles. Furthermore, the findings are grounded in data from the All of Us
Research Program, which, while demographically diverse, does not fully capture population
variation outside the United States.
The digital twins presented in this case study are static, reflecting averaged cluster-level
values rather than individual, real-time updated personal health data. This limits its use in
personalized health monitoring or clinical decision-making. A true digital twin would involve
real-time integration of personal health data and adaptive modeling Sel et al. (2024). In its
current form, the output functions more accurately as a “digital cousin” than a true digital twin
2.8.1.
70 Case Study 3: Designing Static Digital Twin Prototypes
To enhance user accessibility and engagement, complex health variables were translated
into simple icons and visual cues. While effective for communication, these simplifications
may risk misinterpretation or oversimplification. Some icons (e.g., for parental history of
heart attack) may not be intuitively understood without further explanation.
The content presented in the infographic, including the cluster description, the cluster com-
parison, and the personalized advice, is a data-driven implementation in code by Borkenhagen
Borkenhagen (2025). Clinical guidelines from reputable sources, including the World Health
Organization, informed the thresholds used for lifestyle evaluations and recommendations.
However, the resulting insights should not be considered equivalent to medical advice pro-
vided by a licensed healthcare professional.
The purpose of the infographic is to illustrate the potential of data-driven health commu-
nication, not to replace professional diagnosis or treatment. The cluster comparison highlights
how an individual’s values differ from the average values in their assigned cluster. These dif-
ferences form the basis for generating the personalized advice shown.
A key limitation of this study is the lack of user testing. Due to significant delays in gaining
access to the dataset, there was limited time available for conducting evaluations with end
users. As a result, the usability, clarity, and relevance of the outputs have not yet been validated
from the perspective of patients or the general public. Future work should prioritize user
involvement to ensure that the tools developed are truly effective in real-world communication
settings.
8.8 Discussion
The process of translating complex health data into accessible visual formats highlighted both
the opportunities and limitations of digital twin communication. The integration of human
figure visualizations, health icons, a dashboard, and infographic design demonstrated the po-
tential to engage users in intuitively understanding their health status. At the same time, the
project highlighted the importance of careful design to prevent oversimplification.
A particularly promising aspect of the study was the development of a personalized info-
graphic, which illustrated how a static digital twin could be adapted to reflect an individual’s
health profile and support tailored health advice. This concept has potential for future lifestyle
recommendation infographics.
8.9 Conclusion
This case study demonstrates how large-scale, population-based health data can be trans-
formed into personalized, comprehensible outputs through a combination of cluster analy-
sis, visual design, and multi-audience communication. By leveraging All of Us datasets, the
project employed data analysis, which resulted in the identification of five meaningful health-
related clusters. These clusters were then translated into visual representations, laying the
8.9 Conclusion 71
foundation for a static digital twin.
A key contribution of this study is the development of a multi-modal dashboard and a
personalized infographic prototype that integrates individual health data with population-level
insights. This approach illustrates how complex health metrics can be communicated in an
accessible, engaging, and informative format.
While the resulting model is closer to a digital cousin than a fully individualized digital
twin, it nonetheless offers a prototype for future development. A natural next step would be
the integration of dynamic, real-time data and longitudinal tracking.
This study contributes to a broader mission to enhance health communication for patients
by utilizing visualizations and accessible language. The project sought not only to advance
academic research but also to inform and empower the general public. By providing individu-
als with a visual and data-informed understanding of their health, this work fosters increased
public engagement, enhances health literacy, and promotes preventive lifestyle behaviors.
72 Case Study 3: Designing Static Digital Twin Prototypes
Chapter 9
Discussion
This thesis has examined various forms of visualization designed to enhance accessibility and
understanding of health-related information. The project began with visual representations
tailored to a general audience, as outlined in Chapter 6, where the main objective was to
communicate information in a clear and engaging manner. The work then shifted its focus in
Chapter 7, exploring public attitudes toward climate change using the Quality of Life Survey
(2023). This phase required careful interpretation of data to identify recurring patterns in
climate change worry.
To explore public attitudes towards health-related visualizations, a targeted survey was
conducted. The survey also incorporated several questions concerning digital twins, intended
to provide a foundation for the development of digital twins discussed in Chapter 8.
Building on these foundations, the thesis advanced to a more complex topic in Chapter 8,
which explored the concept of digital twins in healthcare. Despite the complexity of digital
twins, visualization consistently serves as a means of enhancing understanding and promoting
healthy lifestyle choices.
This chapter reflects on the feedback received from users, examines the theoretical prin-
ciples underpinning the visual design approach, and evaluates the broader implications of the
findings.
9.1 Implementation of Design Science Principles
This study is guided by the seven foundational design science research guidelines proposed by
Hevner et al Hevner et al. (2004) 3.1. Each guideline is addressed explicitly in the research
process.
Design as an Artifact
The central outcomes of this research are a series of visual artifacts, including infographics
and dashboards, designed to facilitate the communication of both simple and complex health-
74 Discussion
related information. These artifacts address the topics of dietary guidelines, the role of health,
demographic and attitudinal factors in shaping concern about climate change, and the repre-
sentation of personalized health through digital twins. Each artifact is designed to enhance
user comprehension, foster engagement, and increase accessibility for diverse audiences, in-
cluding the general public, patients, and healthcare professionals.
Problem Relevance
This research is driven by a critical need to enhance and tailor the communication of health-
related information for the benefit of the patients. By focusing on digital twin visualiza-
tions tailored to real-world healthcare challenges, the study contributes to the development
of technology-based solutions that address critical issues in public health literacy. Beyond
complex and emerging health technologies, the study also emphasizes the importance of visu-
alizations for more accessible topics, such as dietary guidelines. This aspect of the research
explores whether visual communication can stimulate interest and encourage the adoption of
healthier eating behaviors. Such a focus is particularly significant given the well-documented
relationship between diet and cardiovascular health Barbaresko et al. (2018); World Health
Organization and Food and Agriculture Organization of the United Nations (2003).
Design Evaluation
To evaluate the utility, quality, and effectiveness of the visualizations, the study applies both
qualitative and quantitative methods, including focus groups and a survey. These evaluation
strategies played distinct roles across the three case studies: they informed iterative design re-
finements in Case Study 1, validated the overall impact of the visual artifacts in Case Study 2,
and assessed user engagement with the content presented in Case Study 3. This comprehen-
sive evaluation strategy provided nuanced insights into how the visualizations were perceived,
interpreted, and utilized by different target audiences.
Research Contributions
This thesis offers several contributions to the field of health communication, with particular
emphasis on a multi-audience approach. It demonstrates how visualization techniques can be
effectively employed to convey abstract and technically complex health concepts to diverse
user groups. These contributions are situated within the broader domain of artifact develop-
ment and contribute to advancing both theoretical and practical understanding in health com-
munication design.
Research Rigor
9.2 Answering Research Questions 75
This study has applied rigorous research methods throughout its course. The design process
drew on well-established theories from visual communication and user-centered design. To
maintain high standards, the infographic designs in all case studies were evaluated using the
GRAPHIC Principles of Public Health Infographic Design, while the dashboards were evalu-
ated following Bach et al.’s dashboard design principles Bach et al. (2022); Stones and Gent
(2014). These guidelines helped ensure that the visualizations were both effective and aligned
with best practices in health communication.
Design as a Search Process
The development of effective visualizations is conceptualized as a search process that involves
identifying appropriate design strategies to achieve the desired communication outcomes. This
process includes prototyping, user feedback integration (Case study 1), and context-specific
adaptation to ensure alignment with the complex and evolving requirements of the healthcare
domain. Cooperation with Borkenhagen for Case Study 3 was also crucial for exploring the
domain of digital twins Borkenhagen (2025).
Communication of Research
The findings of this research have been communicated through research papers and an aca-
demic lecture. Tailoring Infographics of the Norwegian National Dietary Guidelines was
presented at ICIMTH 2024 10.4, and Harnessing Visualization to Enhance Digital Twins in
Health Applications was presented at MIE 2025 10.4. Additionally, the paper Visualizing the
Intersection of Climate Concerns, Health, Attitudes, and Demographics has been accepted for
publication in the proceedings of ICIMTH 2025. An academic lecture was also delivered at
the SDG Conference in Bergen Babic et al. (2025). The thesis will also be published in the
University of Bergen’s open-access repository, BORA. As a result, various components of
the thesis have been shared with both academic audiences and will be made available to the
general public.
9.2 Answering Research Questions
In addition to background, methodology, and case study reflections, the survey (10.4) serves
as a key source for addressing the research questions.
RQ 1: How can visualization support understanding of health-related information and
enhance patient engagement?
Visualizations, such as infographics and dashboards, can contribute to the understanding of
health-related information. Infographics are effective tools for simplifying complex medical
content, making it more accessible and easier to comprehend Martin et al. (2019); Piil et al.
76 Discussion
(2023). Dashboards have also been shown to improve care processes and are associated with
better patient outcomes Dowding et al. (2015).
Although Case Study 1 focused on relatively straightforward health information, focus
group participants responded with a high level of engagement. They preferred the infographic
version of the dietary guidelines over the text-based alternative. This preference suggests that
even for basic content, visualization can foster greater interest and involvement.
The survey’s findings further reinforce this observation. When participants were asked how
they best learn medical information, 59% indicated a preference for a combination of visuals
and text, while 38% preferred visualizations alone. These results suggest that people are more
likely to engage with health-related technologies when information is presented in a visually
intuitive way.
By simplifying complex content, visualizations facilitate more informed decision-making
and enable people to build a stronger connection to their health data. This increased clarity
and accessibility have the potential to enhance both understanding and engagement.
RQ 2: How can abstract health concepts be made more understandable for multi-
audience communication?
Communicating effectively with multiple audiences requires deliberate choices in structure,
storytelling, and design. These decisions help ensure that the message remains clear, engaging,
and relevant across different user groups Mörth (2022).
One of the primary objectives of this project has been to make health information more
accessible to a broader audience. In that context, multi-audience communication becomes es-
pecially important. As Mörth points out, this approach reflects a broader shift in visualization
research, where increasing attention is being given to how visualizations are presented and
interpreted by diverse audiences Mörth (2022).
Case Study 3 addresses the abstract health concept of digital twins, and the design pro-
cess included deliberate choices to enhance both accessibility and relevance. The infographic
prototype, in particular, highlighted the importance of balancing visual and textual elements.
It combined figures, icons, and text to communicate complex ideas in a clear and engaging
way. As Mörth emphasizes, multi-audience communication is not merely about simplifying
content; it is about creating content that resonates across levels of expertise Mörth (2022).
Survey responses support this idea. 66% of participants agreed that visualizations improve
the clarity of digital twin data. This suggests that visual tools can play a crucial role in making
abstract health concepts more comprehensible.
RQ 3: Which visualization format is most effective for communicating digital twin mod-
els?
In healthcare, digital twin visualizations can benefit from both infographics and dashboards,
depending on the target audience Andreassen and Babic (2025b). Infographics are particu-
9.2 Answering Research Questions 77
larly effective for communicating with patients, while dashboards are suited for healthcare
professionals who require more detailed and technical information.
Sævareid highlights this dual approach in the context of arthroplasty digital twins Sæ-
vareid (2024). In her study, dashboards were used to present the current patient status, along
with detailed graphs, while infographics conveyed key information, such as recovery times,
complication rates, and overall success rates. This combination provided a comprehensive
overview tailored to meet the diverse needs of different users. The integration of infographics
alongside statistical data not only improved the aesthetic appeal but also enhanced the clarity
and accessibility of the information Sævareid (2024).
A similar distinction is evident in Case Study 3. The multi-modal dashboard prototype
developed for this case may be particularly useful for healthcare professionals, while the info-
graphic prototype is more appropriate for patient communication.
RQ 4: How can digital twin visualizations enhance cardiovascular disease lifestyle
choices?
Cardiovascular diseases (CVD) remain the leading cause of death worldwide, with modifiable
lifestyle factors playing a significant role in their prevalence Barbaresko et al. (2018). Re-
search shows that adopting a greater number of healthy behaviors significantly reduces the
risk of CVD Barbaresko et al. (2018). Digital twin visualizations provide a novel approach
to addressing this issue by making health recommendations more personalized. In the Ba-
sic Archetype Model, which forms the framework for Case Study 3, these visualizations can
illustrate how individuals within a given cluster are performing, how a specific patient com-
pares to others in the same cluster, and what personalized health advice may help improve
lifestyle-related choices.
Digital twin visualizations have the potential to positively influence lifestyle choices re-
lated to cardiovascular disease by making health information more comprehensible and ac-
tionable. The survey revealed that 93% of respondents expressed interest in using digital twins
to track and improve heart health, with many citing personalized care (90%) and better health
interventions (52%) as key benefits.
RQ 5: What are current attitudes towards visualization of Digital Twins?
Digital twins offer significant potential for enhancing proactive healthcare by facilitating the
early detection of medical conditions. This kind of anticipatory care can help improve patient
outcomes and reduce healthcare costs by preventing complications and avoiding unnecessary
hospitalizations (Randles, 2025).
Survey results suggest that public attitudes toward digital twin technology are cautiously
optimistic. Around 41% of respondents said they would be willing to use a digital twin to mon-
itor their heart health, with another 55% open to the idea. When asked about sharing personal
health data to improve outcomes in a digital twin system, 48% said they felt very comfortable,
78 Discussion
31% were neutral, and 21% were uncomfortable. As for how people prefer to receive health
information, only 10% chose digital twin technology alone, 21% preferred traditional methods
like in-person consultations or phone calls, and the majority, 69%, preferred a mix of both.
Still, there are clear concerns. Privacy and trust were identified as major barriers to adop-
tion, with 76% of respondents citing data privacy as a key issue and 55% pointing to a lack
of public trust. While the technology holds considerable promise, building confidence will
require transparent systems and thoughtful implementation.
These concerns are not unfounded. Digital twins rely on rich, often longitudinal personal
data, which cannot be fully anonymized like traditional health records (Banerjee et al., 2024).
Without proper precautions, the use of such data could pose ethical risks. Strict access controls
and ongoing oversight are essential, with clearly defined rules about who can use a digital twin
and for what purpose. If implemented carefully, they can ensure that digital twin research and
applications benefit healthcare without undermining individual rights or public trust (Banerjee
et al., 2024).
Chapter 10
Conclusions and Future Work
This chapter consolidates the key insights from the thesis, outlines its contributions, and re-
flects on some of the challenges and limitations encountered during the research process. It
also suggests directions for further development and research, especially regarding the poten-
tial of visualization within public health and digital health innovation.
10.1 Conclusions
This thesis has investigated the role of visualization in enhancing understanding and engage-
ment with health-related content. Through the development and evaluation of prototypes, the
study demonstrated how visual representations can facilitate users’ engagement with medical
information and support customized health insights.
Across the three case studies, different types of health-related content were visualized. De-
spite differences in scope, all three cases employed similar visualization techniques and design
principles, aiming to investigate the use of these techniques with varying levels of information
difficulty. This supports the feasibility and flexibility of applying a shared methodological
approach across diverse health content domains.
Furthermore, this thesis contributes to the growing importance of adapting visualization
to different levels of health literacy by using a multi-audience approach. By integrating user-
centred design strategies, the study emphasises the importance of tailoring visual content and
interactions to meet the needs and preferences of different user groups. This is particularly im-
portant in healthcare settings, where clarity, comprehension, and trust are essential to informed
decision-making.
10.2 Limitations
Despite its contributions, this study is subject to several limitations. First, the prototype eval-
uations and the survey conducted involved a limited number of participants, which restricts
the generalizability of the findings. Second, Case Study 3 did not include user testing, which
80 Conclusions and Future Work
limits insight into how users might interpret or interact with the digital twin concept in prac-
tice. Third, the participants involved in the evaluations came from a relatively homogeneous
background, which may not accurately reflect the broader diversity of the population. As a re-
sult, the findings may not fully account for the needs or responses of individuals from different
demographic or socio-cultural groups.
10.3 Future Work
Future research should prioritize user testing with broader and more diverse participant groups.
Including individuals from a wider range of demographic and socio-cultural backgrounds will
provide deeper insight into how well the visualizations perform in terms of effectiveness, us-
ability, and understanding across different populations and real-world health communication
contexts. This approach will also address limitations related to the homogeneity of the focus
group and survey participants, making the designs more inclusive.
Further development of the digital twin prototype should include integration of real-time
health data from wearable technologies or electronic health records. This would enable a more
accurate digital twin that accurately reflects the current state of health. It would also use more
of the potential of the All of Us database, which was selected in part due to its extensive
repository of wearable health data. Leveraging this resource more fully could enhance the
digital twin prototype’s capacity to support personalized medicine.
Another promising direction involves the implementation of adaptive visualization tech-
niques. Schreck et al. propose a conceptual framework for consumer health information sys-
tems that dynamically tailors health information to users’ unique health literacy levels and
visualization preferences Schreck et al. (2023). Building upon this framework, future im-
plementations of lifestyle recommendations and digital twin infographics could incorporate
methods for automatically inferring user preferences by accounting for factors such as age,
gender, prior medical knowledge, and cognitive characteristics.
Moreover, the visualizations would benefit from more rigorous and systematic design eval-
uations. Early and sustained engagement with healthcare professionals is also crucial to ensure
that the designs align with the practical demands of healthcare communication.
Finally, future work should examine the ethical and societal implications of visualizing
personal health data. Topics such as privacy, informed consent, and the risk of misinterpreting
complex medical information should be addressed.
10.4 Final Remarks
This thesis set out to explore the role of visualization in making health information more
accessible, understandable, and engaging. The work demonstrates that visualizations have
strong potential to support public health messaging and personalized health communication.
By integrating design science, user-centered design, and domain-specific design principles,
10.4 Final Remarks 81
the project lays a foundation for future developments at the intersection of visualization, and
digital health. With further refinement and evaluation, these tools can play a crucial role in
helping individuals make sense of the increasing volume of health data they encounter in their
daily lives.
82 Conclusions and Future Work
Bibliography
Abudiyab, N. A., and A. T. Alanazi (2022), Visualization techniques in healthcare applica-
tions: A narrative review, Cureus,14(11), e31,355, doi:10.7759/cureus.31355, available
from: https://doi.org/10.7759/cureus.31355. 1, 2.6
All of Us Research Program (2021), About the all of us research program, accessed 2025-03-
19. Available from: https://allofus.nih.gov/article/about. 5.2
All of Us Research Program (2025a), About the research hub, accessed 2025-05-23. Available
from: https://www.researchallofus.org/about-the-research-hub/. 5.2
All of Us Research Program (2025b), All of us adds data from
50% more participants in largest data expansion to date, accessed
2025-04-09. Available from: https://allofus.nih.gov/article/
announcement-all-of-us-adds-data-from-50-more-participants-in-largest-data-expansion-to-date.
5.2
Amri, M. M., and S. A. Abed (2023), The data-driven future of healthcare: A review,
Mesopotamian Journal of Big Data,2023, 70–76, doi:10.58496/MJBD/2023/010, received
19 Apr 2023, Accepted 20 May 2023. 1, 2.7
Andreassen, J. I., and A. Babic (2025a), Tailoring infographics of the norwegian national
dietary guidelines, Studies in Health Technology and Informatics,323, 260–264, doi:
10.3233/SHTI250091, available from: https://doi.org/10.3233/SHTI250091. (doc-
ument), 2.1, 2.4, 6.1, 6.3, 6.3, 6.3, 6.3.1, 6.5, 6.1, 6.4
Andreassen, J. I., and A. Babic (2025b), Harnessing visualization to enhance digital twin in
health applications, Studies in Health Technology and Informatics,327, 595–596, doi:10.
3233/SHTI250409, accessed 2025-04-09. Available from: https://doi.org/10.3233/
SHTI250409. 9.2
Auflem, M. (2023), Development of medical training equipment through prototyping, wayfar-
ing, and triple loop learning: Industrial research at laerdal medical, Ph.D. thesis, Norwegian
University of Science and Technology (NTNU), available from: https://hdl.handle.
net/11250/3072023. 3.6, 6.4.2
84 BIBLIOGRAPHY
Babic, A., J. I. Andreassen, and A. Borkenhagen (2025), Heart health in the age of climate
change: Harnessing digital twins for personalized wellness, academic lecture presented at
the SDG Conference: Sustainable Health in Turbulent Times, Bergen, Norway. Organized
by The National Sustainability Committee. 7.1, 9.1
Bach, B., E. Freeman, A. Abdul-Rahman, C. Turkay, S. Khan, Y. Fan, and M. Chen (2022),
Dashboard design patterns, IEEE Transactions on Visualization and Computer Graphics,
PP, 1–11, doi:10.1109/TVCG.2022.3209448. 1, 2.2, 3.5.2, 7.9, 8.4.3, 9.1
Banerjee, J., A. Novakowski, D. Duglan, and M. Nikolov (2024), How to trust your digital twin
when it comes to improving your health, accessed: 2025-05-27. Available from: https:
//sagebionetworks.pubpub.org/pub/18v6zwbm/release/3. 2.8, 9.2
Barbaresko, J., J. Rienks, and U. Nöthlings (2018), Lifestyle indices and cardiovascular
disease risk: A meta-analysis, American Journal of Preventive Medicine,55(4), 555–
564, doi:10.1016/j.amepre.2018.04.046, accessed: 2025-04-09. Available from: https:
//doi.org/10.1016/j.amepre.2018.04.046. 2.3, 9.1, 9.2
Barstad, A. (2021), Blir vi stadig mer ensomme?, SSB analyse,8, accessed: 2025-
04-09. Available from: https://www.ssb.no/sosiale-forhold-og-kriminalitet/
artikler-og-publikasjoner/blir-vi-stadig-mer-ensomme. 5.1
Barstad, A. (2022), Negative livshendelser – hvem rammes mest?,
SSB analyse,6, accessed: 2025-04-09. Available from: https://
www.ssb.no/sosiale-forhold-og-kriminalitet/levekar/artikler/
negative-livshendelser--hvem-rammes-mest. 5.1
Baskerville, R. (2008), What design science is not, European Journal of Information Systems,
17(5), 441–443. 3.1
Beecher, B., A. O’Doherty, B. Goulao, A. Jalali, J. Salsberg, L. Dore, et al. (2023), Designing
infographics in health research with patients and the public: A scoping review protocol,
PLOS ONE,18(9), e0291,066, doi:10.1371/journal.pone.0291066, accessed 2025-04-09.
Available from: https://doi.org/10.1371/journal.pone.0291066. 1, 2.1
Björnsson, B., C. Borrebaeck, N. Elander, T. Gasslander, D. R. Gawel, M. Gustafsson,
R. Jörnsten, E. J. Lee, X. Li, S. Lilja, D. Martínez-Enguita, A. Matussek, P. Sandström,
S. Schäfer, M. Stenmarker, X. F. Sun, O. Sysoev, H. Zhang, and M. Benson (2020),
Digital twins to personalize medicine, Genome Medicine,12(4), available from: https:
//doi.org/10.1186/s13073-019-0701-3. 2.8
Borkenhagen, A. (2025), A multi-level framework for cardiovascular digital twins: A lifestyle-
centric approach, Master’s thesis, University of Bergen, submitted: 02/06/2025. 1.3, 3.10,
8.1, 8.4.4, 8.7, 9.1
BIBLIOGRAPHY 85
Borkenhagen, A., and A. Babic (2025), Developing lifestyle-focused digital twin archetypes
in heart care, Studies in Health Technology and Informatics,323, 265–269, doi:10.
3233/SHTI250092, accessed: 2025-05-26. Available from: https://doi.org/10.3233/
SHTI250092. 2.8.3, 8.3, 8.6.3
Bush, S., and A. Clayton (2022), Facing change: Gender and climate change attitudes world-
wide, American Political Science Review,117, 1–18, doi:10.1017/S0003055422000752.
2.10
Canva (2024), Canva pro, Software, accessed: 2024-03-25. Available from: https://www.
canva.com/pro/. 4.1
Canva (2025), Canva | linkedin, accessed: 2025-04-09. Available from: https://www.
linkedin.com/company/canva/?originalSubdomain=au. 4.1
Cellina, M., M. Cè, M. Alì, G. Irmici, S. Ibba, E. Caloro, D. Fazzini, G. Oliva, and S. Papa
(2023), Digital twins: The new frontier for personalized medicine?, Applied Sciences,
13(13), accessed: 2025-04-09. Available from: https://www.mdpi.com/2076-3417/13/
13/7940. 2.9
Chain, G. S., B. M. Chain, and F. B. Pelliccia (2022), Climate change affects health: Are we
listening?, Global Pediatric Health,9, doi:10.1177/2333794X221091799, accessed: 2025-
06-02. 2.10, 7.4
Clark, D. (2022), Some big news: Flourish is now part of the canva family, accessed: 2025-
04-09. Available from: https://flourish.studio/blog/flourish-canva/. 4.2
Demuth, S., J. D. Sèze, G. Edan, T. Ziemssen, F. Simon, and P.-A. Gourraud (2025), Digital
representation of patients as medical digital twins: Data-centric viewpoint, JMIR Medical
Informatics,13, e53,542, accessed: 2025-05-10. Available from: https://medinform.
jmir.org/2025/1/e53542. 2.8.3, 8.6.1
Dowding, D., R. Randell, P. Gardner, G. Fitzpatrick, P. Dykes, J. Favela, S. Hamer,
Z. Whitewood-Moores, N. Hardiker, E. Borycki, and L. Currie (2015), Dashboards for im-
proving patient care: Review of the literature, International Journal of Medical Informatics,
84(2), 87–100, doi:10.1016/j.ijmedinf.2014.10.001, accessed: 2025-05-10. Available from:
https://medinform.jmir.org/2025/1/e53542. 1, 2.2, 9.2
Fitzgerald, J., C. Gomes, and P. G. Larsen (Eds.) (2024), The Engineering of Digital Twins,
Springer, Cham, doi:10.1007/978-3-031-66719-0, accessed: 2025-04-09. Available from:
https://link.springer.com/book/10.1007/978-3-031-66719-0. 2.8
Flourish (2025), About us, accessed: 2025-04-09. Available from: https://flourish.
studio/company/. 4.2
86 BIBLIOGRAPHY
Gesicho, M., and A. Babic (2022), Designing a dashboard for hiv-data reporting performance
by facilities: Case study of kenya, Studies in Health Technology and Informatics,295, 238–
241, doi:10.3233/SHTI220706. 2.2
Goetz, L. H., and N. J. Schork (2018), Personalized medicine: motivation, challenges, and
progress, Fertility and Sterility,109(6), 952–963, accessed: 2025-04-09. Available from:
https://www.sciencedirect.com/science/article/pii/S0015028218304072. 2.9
Graham, S., and J. Brookey (2008), Do patients understand?, The Permanente Journal,12(3),
67–69, doi:10.7812/TPP/07-144. 1, 2.5
Grant, R. (2018), Data Visualization: Charts, Maps, and Interactive Graphics, ASA-CRC
Series on Statistical Reasoning in Science and Society, 1st ed., 248 pp., Chapman and Hal-
l/CRC. 1, 2.6
Grossglauser, M., and H. Saner (2014), Data-driven healthcare: from patterns to actions, Eu-
ropean Journal of Preventive Cardiology,21(2S), 14–17, doi:10.1177/2047487314552755,
accepted 4 September 2014. 1, 2.7
Helsedirektoratet (2017), Hva gjør helsedirektoratet?, https://www.helsedirektoratet.
no/om-oss/hva-gjor-helsedirektoratet, updated 2024 Apr 1. Cited 2024 Oct 22.
6.2, 6.3.1
Hernandez-Sanchez, S., V. Moreno-Perez, J. Garcia-Campos, J. Marco-Lledó, E. M.
Navarrete-Muñoz, and C. Lozano-Quijada (2021), Twelve tips to make successful medical
infographics, Medical Teacher,43(12), 1353–1359, doi:10.1080/0142159X.2020.1855323,
accessed 2025-04-09. Available from: https://doi.org/10.1080/0142159X.2020.
1855323. 1, 2.1
Hevner, A. R., S. T. March, J. Park, and S. Ram (2004), Design science in information systems
research, Management Information Systems Quarterly,28(1), 75–105. (document), 3.1,
3.1.1, 3.1.3, 3.1, 9.1
Hickman, C. (2020), We need to (find a way to) talk about . . . eco-anxiety, Journal of Social
Work Practice,34(4), 411–424, doi:10.1080/02650533.2020.1844166, accessed: 2025-04-
09. Available from: https://doi.org/10.1080/02650533.2020.1844166. 2.10
Hidalgo, M. C., I. Pisano, and A. Moorby (2010), Determinants of risk perception and
willingness to tackle climate change. a pilot study, PsyEcology,1(1), 105–112, doi:
10.1174/217119710790709595, accessed: 2025-04-09. Available from: https://doi.
org/10.1174/217119710790709595. 2.10, 7.5
Hodgkinson, J. H., and M. Elmouttie (2020), Cousins, siblings and twins: A review
of the geological model’s place in the digital mine, Resources,9(3), 24, doi:10.3390/
BIBLIOGRAPHY 87
resources9030024, accessed: 2025-04-09. Available from: https://doi.org/10.3390/
resources9030024. 2.8.1
IBM (2025), Ibm spss software, accessed: 2025-04-09. Available from: https://www.ibm.
com/spss. 4.3, 7.2
Johannesson, P., and E. Perjons (2021), An Introduction to Design Science, 2nd ed. ed.,
Springer. 3.1, 3.1.2, 3.5, 3.6
Martin, L. J., A. Turnquist, B. Groot, S. Y. M. Huang, E. Kok, B. Thoma, and J. J. G. van
Merriënboer (2019), Exploring the role of infographics for summarizing medical literature,
Health Professions Education,5(1), 10, cited 2024 Oct 25. Available from: https://doi.
org/10.1016/j.hpe.2018.03.005. 2.1, 9.2
Max Planck Institute for Intelligent Systems (2013), Bmi visualizer, accessed: 2025-05-24.
Available from: https://www.bmivisualizer.com. (document), 8.4.1, 8.2
McCrorie, A. D., C. Donnelly, and K. J. McGlade (2016), Infographics: Healthcare communi-
cation for the digital age, The Ulster Medical Journal,85(2), 71–75, accessed 2025-04-09.
Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4920488/. 1, 2.1
Mörth, E. (2022), Scaling up medical visualization: Multi-modal, multi-patient, and multi-
audience approaches for medical data exploration, analysis and communication, Phd thesis,
University of Bergen, Norway. 1, 3.9, 9.2
National Heart, Lung, and Blood Institute (2025), Calculate your bmi, accessed: 2025-05-27.
Available from: https://www.nhlbi.nih.gov/calculate-your-bmi. 8.3
Norwegian Directorate of Health (2025a), Norwegian directorate of health, https://www.
helsedirektoratet.no/en, accessed: 2025-02-22. 5.1
Norwegian Directorate of Health (2025b), Kostråd – verktøykasse (nutrition guidelines
toolkit), accessed: 2025-03-27. 6.5
NTB (2024), Nye kostråd: Helsedirektoratet vurderer influensere for
å nå unge, Dagens Medisin, cited 2024 Oct 22. Available from:
https://www.dagensmedisin.no/helsedirektoratet-kosthold/
nye-kostrad-helsedirektoratet-vurderer-influensere-for-a-na-unge/
646971. 6.2
OpenAI (2025), Chatgpt (gpt-4), accessed: 2025-03-09. Available from: https://chat.
openai.com. 4.4
Perez, S. (2013), Backed by $3m in funding, canva launches
a graphic design platform anyone can use, accessed: 2025-
04-09. Available from: https://techcrunch.com/2013/08/26/
88 BIBLIOGRAPHY
backed-by-3-million-in-funding-canva-launches-a-graphic-design-platform-anyone-can-use/.
4.1
Piil, K., P. Pedersen, H. H. Gyldenvang, A. J. Elsborg, A. B. Skaarup, M. Starklint, T. Kjølsen,
and H. Pappot (2023), The development of medical infographics to raise symptom aware-
ness and promote communication to patients with cancer: A co-creation study, PEC Inno-
vation,2, 100,146, cited 2024 Oct 25. Available from: https://doi.org/10.1016/j.
pecinn.2023.100146. 2.1, 9.2
Poushter, J., M. Fagan, and S. Gubbala (2022), Climate change remains top global
threat across 19-country survey, https://www.pewresearch.org/global/2022/08/31/
climate-change-remains-top-global-threat-across-19-country-survey/, ac-
cessed: 2025-04-09. 2.10, 7.3
Randles, A. (2025), Digital twins in healthcare: Revolutionizing patient care at
duke, accessed: 2025-05-27. Available from: https://comphealth.duke.edu/
digital-twins-in-healthcare-revolutionizing-patient-care-at-duke/. 2.8,
9.2
Røise, W. (2024), Leveraging digital twins for clinical pathways: Exploring arthroplasty reg-
istry and clinical database, Master’s thesis, University of Bergen, Norway. 2.8.2, 8.3
Sahlgaard, C. O. K. (2024), Developing digital twins for predictive outcome analysis in arthro-
plasty, Master’s thesis, University of Bergen, Norway. 2.8.2, 8.3
Schreck, T., D. Albert, M. A. Bedek, K. Horvath, K. Jeitler, B. Kubicek, T. Semlitsch,
L. Shao, and A. Siebenhofer-Kroitzsch (2023), Adaptive visualization of health informa-
tion based on cognitive psychology: Scenarios, concepts, and research opportunities, in
Visualization Psychology, pp. 165–195, Springer, doi:10.1007/978-3-031-34738-2_7, ac-
cessed: 2025-05-27. Available from: https://link.springer.com/chapter/10.1007/
978-3-031-34738-2_7. 10.3
Scikit-learn (), K-means clustering and the silhouette score, accessed: 2025-05-
23. Available from: https://scikit-learn.org/stable/auto_examples/cluster/
plot_kmeans_silhouette_analysis.html. 8.3
Sel, K., D. Osman, F. Zare, S. Masoumi Shahrbabak, L. Brattain, J. Hahn, O. T. Inan,
R. Mukkamala, J. Palmer, D. Paydarfar, R. I. Pettigrew, A. A. Quyyumi, B. Telfer, and
R. Jafari (2024), Building digital twins for cardiovascular health: From principles to clin-
ical impact, European Journal of Information Systems,13(19), available from: https:
//doi.org/10.1161/JAHA.123.031981. 2.8, 8.7
Sharp, H., Y. Rogers, and J. Preece (2019), Interaction Design: Beyond Human-Computer
Interaction, 5 ed., John Wiley & Sons. 3.5, 3.6, 3.7, 3.8.1, 3.8.2, 6.4.1, 6.4.2
BIBLIOGRAPHY 89
Sikt (2023), Surveybanken: Quality of life survey 2023, accessed: 2025-
03-19. Available from: https://surveybanken.sikt.no/no/study/
9ed0f9ca-0de0-4214-8029-223e84b1ee6c/8?type=studyMetadata&file=
71e43223-a95b-43c1-9aa3-2484d7939624/3. 5.1
Sikt (2025), Survey bank (surveybanken), accessed: 2025-04-09. Available from: https:
//surveybanken.sikt.no/. 5.1
Statistisk sentralbyrå (2025), Statistisk sentralbyrå, accessed: 2025-02-22. Available from:
https://www.ssb.no. 5.1, 7.11
Statistisk sentralbyrå, S. Grimstad, and K. S. Støren (2024), Livskvalitetsundersøkelsen 2023,
doi:10.18712/NSD-NSD3173-V1, accessed: 2025-06-02. Available from: https://doi.
org/10.18712/NSD-NSD3173-V1. 5.1
Stones, C., and M. Gent (2014), The 7 graphic principles of infographic design,
Available from: https://visualisinghealth.com/wp-content/uploads/2014/12/
guidelines.pdf, accessed 2025-04-09. 3.5.1, 6.4.3, 7.8.1, 8.4.5, 9.1
Sun, T., X. He, X. Song, L. Shu, and Z. Li (2022), The digital twin in medicine: A key to
the future of healthcare?, Frontiers in Medicine,9, available from: https://doi.org/10.
3389/fmed.2022.907066. 1, 2.8
Sun, T., X. He, and Z. Li (2023), Digital twin in healthcare: Recent updates and challenges,
Digital Health,9, available from: https://doi.org/10.1177/20552076221149651. 2.8
SurveyXact (2025), Surveyxact, https://rambollxact.com/surveyxact, accessed: 2025-
04-09. 4.5
Sævareid, I. W. (2024), Developing conceptual designs for digital twins in arthroplasty, Mas-
ter’s thesis, University of Bergen, Norway. 2.8.2, 8.3, 9.2
Tapia, C., N. Sánchez-Gassen, and A. Lundgren (2023), In all fairness: Perceptions of climate
policies and the green transition in the nordic region, Nordregio Report 2023:5, Nordregio,
accessed: 2025-04-09. Available from: https://www.nordregio.org/publications/
in-all-fairness/. 2.10, 7.3
Ugreninov, E., V. Bjørnshagen, J. S. Vedeler, M. S. Folkvord, and C. Finjarn (2023), Botryg-
ghet på leiemarkedet i dyrtida: En diskusjon av boligsosiale virkemidler, Tech. Rep. NOVA
Notat 5/2023, Velferdsforskningsinstituttet NOVA, OsloMet, accessed: 2025-04-09. Avail-
able from: https://hdl.handle.net/11250/3105340. 5.1
van der Valk, H., H. Haße, F. Möller, and B. Otto (2022), Archetypes of digi-
tal twins, Business & Information Systems Engineering,64, 375–391, doi:10.1007/
s12599-021-00727-7, accessed: 2025-05-26. Available from: https://doi.org/10.
1007/s12599-021-00727-7. 2.8.3, 2.8.3, 8.6.2
90 BIBLIOGRAPHY
WebMD (2023), What is waist-to-hip ratio?, accessed: 2025-05-27. Available from: https:
//www.webmd.com/fitness-exercise/what-is-waist-to-hip-ratio. 8.3
World Health Organization, and Food and Agriculture Organization of the United Nations
(2003), Diet, Nutrition, and the Prevention of Chronic Diseases: Report of a Joint WHO/-
FAO Expert Consultation, World Health Organization, Geneva, wHO Technical Report Se-
ries, No. 916. 2.4, 6.2, 9.1
Appendix
A. Sikt Approval
12.05.2025, 14:18Meldeskjema for behandling av personopplysninger
Side 1 av 5https://meldeskjema.sikt.no/67504ca8-82ab-4cac-b06a-20ebc53bd559/eksport
Notification Form for personal data
Reference number
642582
Which personal data will be processed?
Name
Date of birth
Contact information
Online identifiers
Voice on audio recordings
Other personal information
Describe the other types of personal data
Demografisk data, informasjon om arbeidssituasjon og livsstil
Data controller
Institution responsible for the project
Universitetet i Bergen / Det samfunnsvitenskapelige fakultet / Institutt for informasjons- og medievitenskap
Project leader
Ankica Babic, ankica.babic@uib.no, tlf: +4755589139
Do multiple institutions share responsibility (joint data controllers)?
No
Project information
Title
Visualization of Digital Twins in Heart Health
Summary
Digital twins are a new way to improve healthcare by simulating patient conditions, organs, and treatments in real time. My project
will focus on solving the challenges of showing these digital twins in a clear and useful way. I will work on two types of tools:
dashboards and infographics. Dashboards will be aimed at healthcare professionals and researchers, giving them interactive and
detailed views of digital twin data. Infographics will be designed for the general public, providing simple and clear summaries of
health information. I will also explore basic principles for creating these tools and how they could be used for more than just
professionals. For example, infographics could help patients understand their conditions, and dashboards might one day be used in
doctor visits to show real-time details about a patient’s health. The goal of this project is to make digital twins easier to use and
understand for everyone.
Academic level
Master’s
Contact information, student
Johanne Irene Andreassen, johanneirenea@gmail.com, tlf: 99509131
Primary purpose of the processing
Research or studentpaper
12.05.2025, 14:18Meldeskjema for behandling av personopplysninger
Side 2 av 5https://meldeskjema.sikt.no/67504ca8-82ab-4cac-b06a-20ebc53bd559/eksport
Describe what you will research and why it is necessary to process personal data to achieve the objective
We aim to research the factors influencing heart health and model these using digital twin technology. To achieve this, we will work
with personal data, including lifestyle information, as these factors play a critical role in heart health. Our goal is to create digital twin
visualizations that make the concept relatable and understandable for the general public while maintaining scientific accuracy. The
results of our research will be evaluated by IT and healthcare experts, as well as the general public. We plan to organize a workshop
during the SDG Conference in Bergen to gather feedback on how people perceive digital twins, understand the visualizations, and
identify whether they can relate to a digital twin based on their lifestyle. Our work will leverage databases from Sikt’s Surveybanken
to structure domain knowledge and incorporate findings from literature to validate the results. Additionally, we will use data from
wearables, accessed through the *All of Us* database, to define archetypes and create parallel visualizations. To engage the public,
we will develop a questionnaire tool to collect data and present tailored visualizations, ensuring the results are both accessible and
relevant to diverse audiences.
Total number of data subjects in the project
100-999
Sample 1
Describe the sample
General public, from ages 18-100.
Describe how you will identify or contact the sample
Workshop we are hosting at SDG Conference.
Age group
18 - 100
Which data relating to sample 1 will be processed?
Date of birth
Online identifiers
Other personal information
How will data relating to sample 1 be collected?
Online survey
Summarize the thematic content of the questionnaires and how the they are distributed
The survey will cover topic such as lifestyle, preferences of visualizations and understanding of digital twins. The participants will be
encouraged to find their own digital twin. The survey will be distributed via QR-code, during the SDG conference and on the website
of the University of Bergen.
Attachment
Questionnaire.docx
Legal basis for processing general categories of personal data (GDPR Article 6)
Consent
Information for sample 1
Will the sample receive information about the processing of personal data?
Yes
How does the sample receive information about the processing?
Written (on paper or electronically)
Information letter
Samtykkeskjema_U1.docx
12.05.2025, 14:18Meldeskjema for behandling av personopplysninger
Side 3 av 5https://meldeskjema.sikt.no/67504ca8-82ab-4cac-b06a-20ebc53bd559/eksport
Sample 2
Describe the sample
IT and Clinical Experts
Describe how you will identify or contact the sample
From the relations of our Professor, Ankica Babic, and other professional networks.
Age group
25 - 80
Are any of these groups included in the sample?
Persons residing in countries outside the EU/EEA
Which data relating to sample 2 will be processed?
Name
Date of birth
Contact information
Online identifiers
Voice on audio recordings
Other personal information
How will data relating to sample 2 be collected?
Personal interview
Summarize the thematic content of the interviews and how they are practically conducted
We will seek expertise from the IT field to guide the presentation of digital twins, focusing on applied IT solutions, visualization
techniques, and interactivity. Additionally, we will request feedback on the modeling approaches and the methods used to create the
digital twins. Clinical experts will be consulted to evaluate the digital twins' content, relevance to clinical work, and their potential
applications. This includes assessing their appeal to patients and their usefulness for educational purposes, as well as their ability to
effectively convey information to both professionals and the public.
Attachment
Interview Guide.docx
Legal basis for processing general categories of personal data (GDPR Article 6)
Consent
Information for sample 2
Will the sample receive information about the processing of personal data?
Yes
How does the sample receive information about the processing?
Written (on paper or electronically)
Information letter
Samtykkeskjema_U2.docx
Third persons
Will the project collect information about third persons?
No
Documentation
12.05.2025, 14:18Meldeskjema for behandling av personopplysninger
Side 4 av 5https://meldeskjema.sikt.no/67504ca8-82ab-4cac-b06a-20ebc53bd559/eksport
How will consent be documented?
Manually (on paper)
Electronically (email, e-form, digital signature)
How can consent be withdrawn?
Contact information for the data collectors will be provided to all participants, allowing them to reach out if they wish to withdraw.
How can data subjects get access to their personal data or have their personal data corrected or deleted?
A participant in the project whose information has been collected can contact the data collector, Johanne Irene Andreassen, or the
Data Controller, Ankica Babic, via email or phone to request access, rectification, or deletion of their data.
Approvals
Will any of the following approvals or permits be obtained?
Ikke utfyllt
Security measures
Will directly identifiable data be stored separately from other data ?
Yes
Which technical and practical measures will be used to secure the personal data?
Continuous anonymisation
Encrypted storage
Restricted access
Record of changes
Multi-factor authentication
Access log
Where will the personal data be processed
Mobile devices
Hardware
Describe the data flow and how the data is secured
The data will be stored and collected own the institutions computers and secure cloud services. Written informed consent will be
kept under the lock at the institutions.
Recipients
Who has access to the personal data?
Project leader
Student (student project)
Internal co-workers
Will personal data be transferred to a third country?
No
End of project
Project period
01.02.2025 - 01.07.2025
What happens to the data at the end of the project?
All data will be deleted (deleting raw data)
12.05.2025, 14:18Meldeskjema for behandling av personopplysninger
Side 5 av 5https://meldeskjema.sikt.no/67504ca8-82ab-4cac-b06a-20ebc53bd559/eksport
Will the data subjects be identifiable in publications?
No
Additional information
97
B. All of Us Certificate - Responsible Conduct of Research
AllOfUs Certificate - Responsible Conduct of Research
C. All of Us Certificate - Controlled Tier Data
AllOfUs Certificate - Controlled Tier Data
98 Appendix
D. Related Publications - Full Paper ICIMTH 2024
Tailoring Infographics of the Norwegian
National Dietary Guidelines
Johanne Irene ANDREASSENa,1 and Ankica BABICa,b
a Department of Information Science and Media Studies, University of Bergen, Norway
b Department of Biomedical Engineering, University of Linköping, Sweden
ORCiD ID: Johanne Irene Andreassen https://orcid.org/0009-0003-9022-9668, Ankica
Babic https://orcid.org/0000-0002-7532-6828
Abstract. This research explores how infographics can enhance public
understanding of nutrition based on the Norwegian Directorate of Health’s dietary
guidelines. Although these guidelines are designed for the entire population, the
Directorate has indicated to target younger Norwegians in their outreach efforts. To
make the guidelines more accessible across age groups, we transformed the
Directorate's text-based content into engaging infographics. Three mid-fidelity
prototypes were developed: Infographic 1 emphasizes simplicity with basic
illustrations and structured text; Infographic 2 uses a colorful segmented wheel; and
Infographic 3 combines detailed illustrations with unboxed text. These designs were
evaluated by 72 participants in age-stratified focus groups. Results showed a
preference for Infographics 1 and 3, valued for their clarity and visual appeal, with
age and gender influencing preferences. Based on participant feedback, revised
versions of the infographics were created. The findings confirm that well-designed
infographics can significantly boost engagement, interaction, and retention of
nutritional information.
Keywords. Infographics, Dietary guidelines, Health literacy, Focus group study
1. Introduction
This paper explores the potential of infographics as an effective medium for
communicating nutrition guidelines set forth by the Norwegian Directorate of Health, a
government agency under the Ministry of Health and Care Services dedicated to
improving the health and well-being of Norway’s population while ensuring high
healthcare standards [1].
In August this year (2024), the Directorate published its recent nutrition advice,
targeting the entire population of Norway. Although they express interest in campaigns
that could target all age groups, limited resources mean that efforts are currently
concentrated only on engaging youth audiences [2]. In contrast, older Norwegians
without dedicated campaigns rely solely on text versions of the guidelines, which may
limit learning and retention. This situation presents an opportunity to examine whether
infographics could serve as a cost-effective, accessible learning tool suitable for greater
parts of the population.
1 Corresponding Author: Johanne Irene Andreassen, johanneirenea@gmail.com
Envisioning the Future of Health Informatics and Digital Health
J. Mantas et al. (Eds.)
© 2025 The Authors.
This article is published online with Open Access by IOS Press and distributed under the terms
of the Creative Commons Attribution Non-Commercial License 4.0 (CC BY-NC 4.0).
doi:10.3233/SHTI250091
260
Infographics, which integrate data, visuals, and text, are especially effective in
presenting complex information in an accessible and engaging format [3]. In healthcare,
infographics can enhance information accessibility, aiding the public in understanding
and retaining critical health information [4]. Considering the varied literacy levels and
complexities of health information, infographics can facilitate comprehension by
emphasizing clarity, accessibility, and emotional engagement. Infographics based on the
Norwegian Directorate of Health’s guidelines could thus foster informed decision-
making. This paper focuses on a user-centered approach in infographic design to enhance
user experience by tailoring visuals for the diverse needs of different age groups.
2. Material and Methods
The Norwegian national dietary guidelines of 2024 were utilized to investigate the
impact of infographic design by presenting the same information through three distinct
mid-fidelity prototypes. It cannot be assumed that one infographic can suit the whole
public, therefore we experimented with different visual focuses. The infographic
prototypes were created using Canva Pro [5].
Infographic 1 emphasizes simplicity and clarity, using both basic food illustrations
and text organized within structured text boxes. This layout is designed to present the
text points in a straightforward, accessible way, which can make it suitable for readers
who prioritize quick comprehension of information.
Infographic 2 takes a more visually dynamic approach by arranging the text in a
colorful, segmented wheel resembling a serving plate without food illustrations. Each
segment of the wheel is color-coded, with text formatted in a circular arrangement to
engage readers through visual cues. This infographic may appeal to those who appreciate
an organized yet visually stimulating display, though the layout may challenge readers
accustomed to a more linear structure.
Infographic 3 is designed to attract attention through its rich and detailed food
illustrations, foregoing text boxes entirely. The imagery is intended to immerse and
visually engage readers, creating an aesthetic-focused experience that may appeal to
those who prioritize visuals over structure.
These prototypes were evaluated in focus group sessions stratified by age groups
(10-29, 30-49, 50-69 and 70-89). Participants (n=72) were invited through personal
contacts and email communications, mainly from the University of Bergen. Focus groups
were chosen to capture diverse perspectives, providing both qualitative insights and
quantitative data on preferences for infographic style and structure [6]. The group
sessions were conducted using the prototypes and the text format of the dietary guidelines
as presented in the Directorate’s web page [1], with the organizer facilitating discussions
and taking detailed notes on participants' feedback.
Figure 1. Infographic prototypes 1, 2 and 3 with content translated to English.
J.I. Andreassen and A. Babic / Tailoring Infographics 261
3. Results
The results included both qualitative and quantitative feedback. All participants (n=72)
preferred infographics over plain text. They ranked their favorite infographics and
explained their choices. Quantitative data are shown in Figure 2 as percentage
distributions, highlighting which infographic (1, 2, or 3) was favored by each age group,
segmented further by gender. The percentages are color-coded: Infographic 1 is blue,
Infographic 2 is yellow, and Infographic 3 is red. Notable comments and suggestions
from the focus groups are detailed in Table 1.
Figure 2. Percentage of participants who chose infographic prototypes 1, 2, and 3 as their favorites.
Table 1. Results of the focus group qualitative analysis.
Age groups
Infographic 1
Infographic 2
Infographic 3
10-29 years
(n=40)
Highlighted for its
“great balance
between illustrations
and text,” which
promoted
engagement.
However, one
participant noted it
felt “too
conventional and
boxed-in.”
Not favored as a top
choice. Participants
mentioned missing
illustrations and disliked
the text background color.
Positive feedback
included its ability to
“divide and categorize
guidelines” and being
“the most scientific and
easy to understand.”
Appreciated for “appealing
illustrations” and being
“engaging without much
reading.” Criticized for lacking
structured content. Many found
it the most visually pleasing
infographic, but desired better
organization.
30-49 years
(n=16)
Praised for being the
“best structured,”
with a “good
combination of text
and data” and a “nice
overview.”
Described as “the
most ordered” and
visually appealing.
Mixed feedback, one
participant liked "the
structure and clear
messaging", but felt it
lacked illustrations.
Suggested adding food
images, possibly in the
background.
Praised for its “fresh message”
and visual appeal. One
participant described it as
“ordered, but not too much,”
while another highlighted its
“fine combination of text and
illustrations.” Some found it “a
bit busy, but not overly
disorganized.” One participant
likened it to a “mind map,”
appreciating how it “best
J.I. Andreassen and A. Babic / Tailoring Infographics262
combines illustrations and
information.”
50-69 years
(n = 12)
Noted for
“seamlessly
combining text and
visuals,” making it
easy to understand.
One noted that “the
color choice of green
background fits the
theme of eating
healthy.”
Seen by some as having a
“professional feel.”
Participants mentioned it
“facilitated categorization
of dietary advice” and
“provided a sense of
competence and
authority.” However,
several mentioned that
they missed illustrations.
Reactions were mixed. Some
felt it needed no changes, while
one called it “old-fashioned.”
Criticisms included the
“illustrations in the corner” and
“food images of milk in glass
bottles and donuts and
macaroons as sweets does not fit
Norwegian associations.”
70-89 years
(n = 4)
Viewed as “well-
organized and easy
to follow,” aiding
memory retention of
dietary guidelines.
Not favored due to being
difficult to read, with one
participant noting that
“this was the only
infographic they needed
glasses to read.”
Liked for the illustrations but
criticized for the lack of order.
One saying, “the food in the
illustrations looks so good, I
was motivated to eat healthier
right away!”
Based on the feedback from the focus groups, Infographic 1 stayed the same, while
updated versions of infographic prototypes 2 and 3 were created to better align with user
preferences. For Infographic 2, the changes included adding illustrations to address the
desire for more visual elements, lightening the color scheme to make it more appealing,
and removing the text background to enhance readability. This responded to participants'
feedback on the need for clearer and more engaging visuals. For Infographic 3,
adjustments focused on restructuring the content to improve organization, drawing
inspiration from the balanced layout of Infographic 1. Additionally, the illustrations were
updated to include culturally relevant images of milk and sweets that better fit Norwegian
associations, addressing specific participant comments on visual accuracy and cultural
representation.
Figure 3. New versions of Infographic prototypes 2 and 3 with content translated to English.
4. Discussion
Presenting dietary information to the public is critical, given the relevance of nutrition to
public health. Nutrition is emerging as a key modifiable factor in chronic disease, with
growing evidence that dietary changes significantly impact health [7].
In designing three distinct infographic prototypes, we aimed to understand how
different age groups respond to varied presentation formats. We used Canva Pro [5] to
develop prototypes that explored different combinations of text and visuals. This
approach allowed us to test key design concepts without requiring professional design
J.I. Andreassen and A. Babic / Tailoring Infographics 263
skills. Notably, all participants, regardless of age, could express and justify their
preferences, indicating that our prototypes were intuitive and effectively conveyed the
information. This study suggests that infographics can enhance health literacy, indicating
that clear communication practices can benefit all individuals, not just those with limited
health literacy [8]. However, one prototype will not suit all.
This project lays the groundwork for future infographics on more complex topics,
such as lifestyle choices, adaptable across data types and audiences. Our findings
highlight infographics as powerful tools in health education, with technology enabling
clear, impactful visuals so researchers can focus on content.
5. Conclusion
This study underscores the potential of infographics as a compelling method for
conveying dietary guidelines. Through tailored designs that align with varying
preferences and age groups, infographics can improve health literacy and promote
healthier food choices. By understanding and leveraging the strengths of different
formats, diverse audiences can be engaged effectively, facilitating informed decision-
making regarding nutrition. Our findings suggest to create multiple infographic layouts
to balance text and illustrations according to age-specific preferences. With this
approach, even complex information can be presented effectively.
References
[1] Helsedirektoratet. Hva gjør Helsedirektoratet? [Internet]. Oslo: Helsedirektoratet; 2017 [updated 2024
Apr 1; cited 2024 Oct 22]. Available from: https://www.helsedirektoratet.no/om-oss/hva-gjor-
helsedirektoratet
[2] NTB. Nye kostråd: Helsedirektoratet vurderer influensere for å nå unge [Internet]. Dagens Medisin; 2024
Aug 14 [cited 2024 Oct 22]. Available from: https://www.dagensmedisin.no/helsedirektoratet-
kosthold/nye-kostrad-helsedirektoratet-vurderer-influensere-for-a-na-unge/646971
[3] Martin LJ, Turnquist A, Groot B, Huang SYM, Kok E, Thoma B, van Merriënboer JJG. Exploring the
role of infographics for summarizing medical literature. Health Prof Educ [Internet]. 2019 Mar 1 [cited
2024 Oct 25];5(1):10. Available from: https://doi.org/10.1016/j.hpe.2018.03.005
[4] Piil K, Pedersen P, Gyldenvang HH, Elsborg AJ, Skaarup AB, Starklint M, Kjølsen T, Pappot H. The
development of medical infographics to raise symptom awareness and promote communication to
patients with cancer: A co-creation study. PEC Innov [Internet]. 2023 Dec [cited 2024 Oct 25];2:100146.
Available from: https://doi.org/10.1016/j.pecinn.2023.100146.
[5] Canva Pro [Internet]. Sydney (Australia): Canva; 2024 [cited 2024 Oct 24]. Available from:
https://www.canva.com.
[6] Sharp H, Preece J, Rogers Y. Interaction Design: Beyond Human-Computer Interaction. 5th ed.
Indianapolis (IN): Wiley; 2019. 301 p.
[7] World Health Organization, Food and Agriculture Organization of the United Nations. Diet, nutrition,
and the prevention of chronic diseases: Report of a joint WHO/FAO expert consultation. Geneva: WHO;
2003.
[8] Graham S, Brookey J. Do patients understand? Perm J [Internet]. 2008 [cited 2024 Oct 25];12(3):67-9.
Available from: https://doi.org/10.7812/TPP/07-144
J.I. Andreassen and A. Babic / Tailoring Infographics264
104 Appendix
E. Related Publications - Short Communication Paper MIE
2025
Harnessing Visualization to Enhance
Digital Twin in Health Applications
Johanne Irene ANDREASSENa,1 and Ankica BABIC a,b
a Department of Information Science and Media Studies, University of Bergen, Norway
b Department of Biomedical Engineering, University of Linköping, Sweden
ORCiD ID: Johanne Irene Andreassen https://orcid.org/0009-0003-9022-9668
Ankica Babic https://orcid.org/0000-0002-7532-6828
Abstract. Digital twins simulate patient conditions and treatment processes in
healthcare, but their interpretation remains a challenge. This study reviews existing
literature on visualization tools, specifically dashboards and infographics, and their
application in digital twin healthcare systems. Dashboards support healthcare
professionals by offering interactive, real-time data, while infographics simplify
complex data for public engagement. Through case studies, the paper highlights the
visualization principles of both tools. The findings suggest that improved
visualization methods are crucial for advancing digital twin adoption in healthcare.
Keywords. Digital twin, Visualization, Dashboard, Infographics, Healthcare
1. Introduction
Visualization is crucial for transforming complex healthcare data into actionable insights,
depending on data integration, the task, and the target audience [1]. This study examines
possibilities of visualization of medical digital twins, virtual replicas of physical systems
that simulate real-world processes, enabling faster, cost-effective testing [2][3]. Their
complexity, however, can hinder interpretation, which could be helped by effective
visualization to bridge gaps between researchers, professionals, and patients [4].
2. Methods
This study combined a literature review and case studies to explore visualization methods
and principles in healthcare. Two case studies were selected from the literature review:
(1) Gesicho and Babic’s HIV dashboard, highlighting interactivity and contextualization
[5], and (2) McCrorie et al.’s infographics, emphasizing storytelling and accessibility [6].
Papers were assessed for adherence to visualization principles, comparing dashboards
and infographics in healthcare and broader digital twin applications.
1 Corresponding Author: Johanne Irene Andreassen; E-mail: johanneirenea@gmail.com.
Intelligent Health Systems – From Technology to Data and Knowledge
E. Andrikopoulou et al. (Eds.)
© 2025 The Authors.
This article is published online with Open Access by IOS Press and distributed under the terms
of the Creative Commons Attribution Non-Commercial License 4.0 (CC BY-NC 4.0).
doi:10.3233/SHTI250409
595
3. Results
Building on the visualization principles identified in the literature review, a comparative
table (Table 1) was developed to highlight key visualization principles.
Table 1. Comparison of visualization principles as applied to case studies.
Visualization principle
Case study 1 (Dashboard) [5]
Case study 2 (Infographic) [6]
Data representation
Uses maps, scatter charts, and line
graphs to represent HIV data
Uses charts, images, and icons for
health data
Interactivity
Interactive, allowing user
exploration through filters
Static, with fixed data and a cohesive
narrative
Real-time data updates
Offers real-time data updates to
monitor HIV facility performance
Represents data at a specific moment
in time
Contextualization
Provides contextual data through
color codes and legends
Uses colors and symbols (e.g., traffic
light system) to add context
Visual Hierarchy
Highlights important data points
using scaling and color saturation
Guides focus using size, color, and
position
4. Discussion and Conclusions
Digital twin visualizations in healthcare could benefit from infographics' static,
narrative-driven visuals, while dashboards can offer interactive, real-time data with color
codes. Both methods have the potential to enhance data accessibility, using visual
hierarchy to guide focus through size, color, and position. Additionally, designing
effective visualizations requires careful consideration of diverse user groups,
necessitating tailored design solutions to meet the needs of different audiences [7].
Dashboards and infographics offer significant potential for communicating digital
twins in healthcare, leveraging visualization principles to engage diverse audiences and
enhance understanding.
References
[1] Bebis G, Boyle R, Parvin B, Koracin D, Kuno Y, Wang J, et al. Advances in Visual Computing. Third
International Symposium, ISVC 2007, Lake Tahoe, NV, USA, Nov 26–28, 2007. Proceedings, Part II.
Berlin: Springer-Verlag; 2007. Lecture Notes in Computer Science. Vol. 4842, p. 652–661.
[2] Sun T, et al. The Digital Twin in Medicine: A Key to the Future of Healthcare? Front Med (Lausanne).
2022 Jul 14;9: 907066. doi: 10.3389/fmed.2022.907066. PMID: 35911407; PMCID: PMC9330225.
[3] Björnsson B, Borrebaeck C, Elander N, Gasslander T, Gawel DR, Gustafsson M, Jörnsten R, Lee EJ, Li
X, Lilja S, Martínez-Enguita D, Matussek A, Sandström P, Schäfer S, Stenmarker M, Sun XF, Sysoev
O, Zhang H, Benson M; Swedish Digital Twin Consortium. Digital twins to personalize medicine.
Genome Med. 2019 Dec 31;12(1):4. doi: 10.1186/s13073-019-0701-3. PMID: 31892363; PMCID:
PMC6938608.
[4] Jason LA, Glenwick DS. Handbook of Methodological Approaches to Community-Based Research:
Qualitative, Quantitative, and Mixed Methods. 1st ed. New York: Springer; 2015. p. 301. doi:
10.1093/med:psych/9780190243654.001.0001.
[5] Gesicho M, Babic A. Designing a dashboard for HIV-data reporting performance by facilities: Case study
of Kenya. In: Mantas J, Stoicu-Tivadar L, Chronaki CE, Mihalas G, Boyaci A, Hasman A, et al., editors.
Advances in Informatics, Management and Technology in Healthcare. Amsterdam: IOS Press; 2022. doi:
10.3233/SHTI220706.
[6] McCrorie AD, Donnelly C, McGlade KJ. Infographics: Healthcare Communication for the Digital Age.
Ulster Med J. 2016 May;85(2):71–75. PMID: 27601757; PMCID: PMC4920488.
[7] Andreassen JI, Babic A. Tailoring infographics of the Norwegian national dietary guidelines. Presented
at: 22nd International Conference on Informatics, Management, and Technology in Healthcare; 2024 Dec
13–15; Athens, Greece. Forthcoming publication in: Stud Health Technol Inform.
J.I. Andreassen and A. Babic / Harnessing Visualization to Enhance Digital Twin596
107
F. Related Publications - Full Paper ICIMTH 2025
The paper Visualizing the Intersection of Climate Change Concerns, Health, Attitudes and
Demographic Factors has been accepted for publication in the proceedings of International
Conference on Informatics, Management and Technology in Healthcare (ICIMTH) 2025. The
conference take place on 04-06 July 2025 in Athens, Greece, but the proceedings have not yet
been published. Therefore, the full version of the paper cannot be included at this time. The
paper is based on the research presented in Case Study 2 in this thesis and will be available
upon publication 7.
108 Appendix
G. Survey Responses
Male Female Non-binary Other
I don't want to share this information
What is you gender?
0% 25% 50% 75% 100%
50 50
Respondenter
38
18-29 30-39 40-49 50-59 60-69 70-79 80-89
90+ I don't want to share this information
What is you age?
0% 25% 50% 75% 100%
71 513 53 3
Respondenter
38
Very interested Somewhat interested Not interested
No spesific opinion
What do you think of the idea to use
digital twin to track and improve heart
health?
0% 25% 50% 75% 100%
45 48 7
Respondenter
29
Medical records
Data from wearable devices (e.g. smartwatches)
Surveys/questionnaires Demographic data
What type of data would you trust most
when it comes to managing your heart
health through digital tools?
0% 25% 50% 75% 100%
90 59 324
Respondenter
29
Yes Maybe No No spesific opinion
Would you be open to using digital
health technology (such as a digital twin)
to monitor your heart health?
0% 25% 50% 75% 100%
41 55 3
Respondenter
29
Very comfortable Not comfortable Netural
No spesific opinion
How comfortable would you be with
sharing personal health to improve
health outcomes in a digital twin system?
0% 25% 50% 75% 100%
48 21 31
Respondenter
29
Better health interventions More personalized care
No benefits Other
What benefits do you see in using digital
twins to track well-being on a population
level?
0% 25% 50% 75% 100%
52 90 7
Respondenter
29
Yes Maybe No No spesific opinion
If you could track your health using a
digital twin, would it encourage you to
make healthier lifestyle choices?
0% 25% 50% 75% 100%
45 45 37
Respondenter
29
Privacy concerns Technical feasibility Public trust Other
What do you believe could be the
greatest challenge in implementing a
digital twin for heart health and we...
0% 25% 50% 75% 100%
76 38 55
Respondenter
29
Digital twin technology Traditional methods
A combination of both No specific opinion
Would you prefer to receive health
information through digital twin
technology or more traditional metho...
0% 25% 50% 75% 100%
10 21 69
Respondenter
29
Yes Maybe No No spesific opinion
Do you think digital twin visualizations
enhance the clarity and comprehension
of digital twins?
0% 25% 50% 75% 100%
66 24 37
Respondenter
29
Visualizations Text Combination of visualizations and text
No specific opinion
Do you find it easier to learn medical
information through visualizations or
text?
0% 25% 50% 75% 100%
38 59 3
Respondenter
29
Yes Maybe No No spesific opinion
Do you think that a digital twin could
provide valuable insights for early
detection of health risks?
0% 25% 50% 75% 100%
52 38 10
Respondenter
29
Ny Distribuert Noen svar Gjennomført Frafalt
Samlet status
0% 25% 50% 75% 100%
10 21 69
Respondenter
42
