INMR96 Digital Health and Data Analytics

Semester 2 2024/25

Coursework assignment specification

1. Introduction

This module is assessed 100% through this coursework assignment.

You must submit an individual report that applies digital health and data analytics concepts, methods and techniques learned to address healthcare problems. The report should not exceed 20 pages of A4 with a minimum font  size of 10, including tables and  diagrams/illustrations but excluding the declaration on the use of generative AI tools, references and appendices.

The report must be submitted through Turnitin via Blackboard by the specified deadline.

You are expected to develop digital solutions (e.g. digital technologies, data-driven solutions, AI and machine learning, etc) to address healthcare problems. You are expected to:

•    Understanding of the specific problems and unmet needs in healthcare

•    Identification of current  solutions and potential technologies considering existing healthcare settings

•    Design your solution (e.g. digital technologies, data-driven solutions, AI and machine learning, etc) that addresses the healthcare problems

•     Critically evaluate your solution

•    Discuss ethical implications and regulatory aspects of your solution

It is expected that you will explore and discover relevant data, methods, techniques and tools that contribute to your proposed solution.

Use of generative AI tools:

The use of GenAI tools is actively encouraged in completing this assignment to help you develop your skills in the use of such tools and understand how their use can be incorporated into authentic writing tasks. Specifically, you are expected to develop a critical and responsible approach to the use of AI tools in relation to the module learning objectives.

You must include a statement to acknowledge your use of GenAI tools* within the assessment itself. This statement should be written in complete sentences and include the following information:

• Name and version of the GenAI tool used; e.g. ChatGPT-3.5

• Publisher (company that provides the GenAI system); e.g. OpenAI

• URL of the AI tool (if applicable)

• Brief description (single sentence) of the way in which the GenAI tool was used

• Confirmation that the work is your own For example:

I acknowledge the use of ChatGPT 3.5 (OpenAI, https://chat.openai.com/) to generate an outline for background study. I confirm that no output generated by GenAI has been presented as my own work. Note: if you have not used GenAI tools to help with your assessment, you must still include a statement to acknowledge this fact, e.g. I declare that no GenAI tools have been used to produce this work. The misuse of GenAI tools, including the failure to appropriately acknowledge the use of such tools, is considered academic misconduct and carries sanctions, as detailed in the Assessment Handbook. Please also refer to student guidance on Using Generative AI Tools at University and GenAI and University Study

These tools must be used responsibly, e.g.:

- copying and pasting texts generated by such tools will be considered as academic misconduct (plagiarism); you should use these in the same way as you do with other written materials, i.e., use them critically and paraphrase them;

- use of images and diagrams generated by such tools must be acknowledged and clearly indicated in the figure caption;

- consider critically whether the materials generated by AI impacts the quality of the report, e.g., in terms of coherence, standard, and presentation.

2. The Scope

One of the following four options should be selected:

Option 1:  Early Disease Detection and Prevention

Early Disease Detection and Prevention refers to the use of medical screenings, digital health technologies, artificial intelligence (AI), and data analytics to identify diseases at an early stage—before symptoms become severe. The goal is to enable early interventions, reduce healthcare costs, and improve patient outcomes.

Why is Early Detection Important?

•    Better Treatment Outcomes – Diseases caught early are often easier to treat and may even be reversible.

•    Lower Healthcare Costs – Preventing disease progression reduces hospitalizations and expensive treatments.

•    Improved Quality of Life – Early intervention can slow disease progression and maintain a patient’s overall health.

•    Reduced Mortality Rates – Many life-threatening diseases, such as cancer and cardiovascular diseases, have higher survival rates when diagnosed early.

You are expected to design a solution integrating AI and data analytics with digital health tools that enables proactive identification of health risks, facilitating timely interventions.

You may find the following websites interesting or helpful:

•   AI-Powered Breast Cancer Screening: The UK's National Health Service (NHS) is

conducting a large-scale trial using artificial intelligence to improve breast cancer detection. This trial aims to enhance diagnostic accuracy and efficiency in mammogram screenings.

theguardian.com

•   AI-Driven Early Detection of Diabetic Retinopathy: Google has developed an AI system

called Automated Retinal Disease Assessment (ARDA) to assist in the early detection of diabetic retinopathy, a leading cause of blindness among individuals with diabetes. This AI tool analyzes retinal images to identify disease presence, facilitating timely intervention.

health.google

•   Wearable Devices for Heart Disease Detection: Wearable devices like the Apple Watch

have integrated features to monitor heart rhythms and can detect irregularities such as atrial fibrillation (AFib). Early detection of AFib can prompt users to seek medical advice before more severe symptoms develop.Apple Watch Healthcare Applications

Option 2: Personalized Treatment and Precision Medicine

Personalized Treatment and Precision Medicine are approaches to healthcare that tailor medical decisions, treatments, and interventions to individual patients based on their unique characteristics. These approaches consider factors such as genetics, lifestyle, environment, and medical history to develop the most effective treatment plans.

How it differs from traditional medicine

•    Traditional Medicine: Uses a "one-size-fits-all" approach where treatments are based on average responses across large populations.

•    Personalized/Precision Medicine: Uses biological, genetic, and digital health data to identify treatments that work best for each patient.

You are expected to design digital health/AI/data analytics-based solutions to inform. personalised treatment and precision medicine approaches.

You may find the following websites interesting or helpful:

•   AI-powered personalised medicine could revolutionise healthcare:

https://www.theguardian.com/commentisfree/2023/jun/26/ai-personalise-medicine-patient- lab-health-diagnosis-cambridge

•   New regulatory thinking is needed for AI-based personalised drug and cell therapies in precision oncology https://www.nature.com/articles/s41698-024-00517-w

•   Precision Medicine, AI, and the Future of Personalised Health Care https://pmc.ncbi.nlm.nih.gov/articles/PMC7877825/

Option 3: Chronic Disease Management and Remote Care

Chronic Disease Management (CDM) and Remote Care involve long-term, proactive healthcare strategies to help individuals with chronic conditions maintain their health, prevent complications, and improve quality of life. Remote care enables healthcare professionals to monitor and support patients outside of traditional hospital settings through telehealth, wearable devices, AI, and data   analytics.

Chronic diseases are long-term conditions that require continuous monitoring  and care. Examples include diabetes, hypertension, chronic obstructive pulmonary disease, heart disease, arthritis, cancer, mental health disorders, and chronic kidney disease. Managing these conditions involves lifestyle modifications,  medication  adherence,  routine  monitoring,   and  timely   interventions to  prevent complications such as strokes, heart attacks, and organ failure.

Remote   care   refers    to delivering    healthcare    services    outside   of   hospitals    and    clinics, using telemedicine, AI-powered tools, wearable devices, and mobile health (mHealth) applications. It allows continuous patient monitoring and reduces unnecessary hospital visits.

You are expected to design digital health, AI, and data analytics-based solutions to support chronic disease management and remote care.

You may find the following websites interesting or helpful:

•   Artificial Intelligence and Its Role in the Management of Chronic Medical Conditions: A Systematic Review https://pmc.ncbi.nlm.nih.gov/articles/PMC10607642/

•   Remote Patient Monitoring (RPM) with Wearables: Real-world Example: Freestyle Libre CGM for Diabetes (Abbott Freestyle. Libre)

•   Shaping the future of chronic disease management: Insights into patient needs for AI-based homecare systems:https://www.sciencedirect.com/science/article/pii/S1386505623003192

Option 4: The healthcare problem proposed by yourself

You may also develop digital solutions (e.g. digital technologies, data-driven solutions, AI and machine learning, etc) to address any other healthcare problems that are out of the scope 1-3.

3. Your solution and report

Your solution must include the following, which should be included in the report.

Title - You should provide a title that states a meaningful title for the proposed solution. In the report, this should not be on a separate page as it will count towards the page limit ; no other information, such as your name, should be included in the report.

1. Introduction - You should provide a brief introduction that outlines the proposal and elements of the report.

2. Problem analysis - You should provide an analysis of the specific healthcare problem to be addressed. Expected contents here include a description of the context, background, and key issues to be addressed. It is suggested to integrate the user-centric design1  into the problem analysis.

3. Current solutions and potential technologies - You are required to conduct research in  sufficient scope and depth. There should be a demonstration of understanding of existing solutions in the relevant problem area or potential technologies that can be adopted to address the problem.

4. Solution design - You are required to provide the digital health and data analytics solution design to address the proposed healthcare problems. A detailed solution design and description are required, e.g. what digital technologies are you using? How do different technologies work together? How is the data generated, managed, used and analysed? How does your solution benefit end-users? It is suggested to integrate the user-centric design method1 into the solution and intervention design.

5. Critical evaluation and discussion- You are expected to present a critical evaluation of your solutions, e.g. cost and infrastructure requirements, user acceptance, limitations and constraints, comparison with existing solutions etc.  It is suggested to integrate the roadmap of responsible machine learning for health care2  into critical evaluation and discussion.

6. Ethical implications- You need to consider any ethical and regulations implications to be considered in your solutions, e.g. patient privacy and safety, data protection, medical device regulations etc.

7. Conclusion - you should provide a brief conclusion that summarises the proposed solution and outlines its key innovative features.

The following also must be provided but are outside the page limit:

Declaration - see above regarding the use of generative AI tools.

References - you should provide a list of references following the Harvard (author-year) referencing format.

Appendices - you may provide appendices as appropriate; those provided as appendices may not be read and are not part of the assessment but can provide additional information and material that supplements the content of the report.

The report should be written professionally and structured with section numbers and headings as specified above. This is reflected in the assessment of the quality of report in terms of consistency, logical argumentation, coherence, flow, style. and structure. Each section may have subsections as appropriate to clarify the structure and flow of the report. There is no need for an abstract or table of contents.

References:

1 Wiens, J., Saria, S., Sendak, M. et al. Do no harm: a roadmap for responsible machine learning for health care. Nat Med 25, 1337—1340 (2019).https://doi.org/10.1038/s41591-019-0548-6

2 Seneviratne, M.G., Li, R.C., Schreier, M., Lopez-Martinez, D., Patel, B.S., Yakubovich, A., Kemp, J.B.,

Loreaux, E., Gamble, P., El-Khoury, K. and Vardoulakis, L., 2022. User-centred design for machine learning in health care: a case study from care management. BMJ Health & Care Informatics, 29(1).

4. Marking scheme

The team presentation is not assessed but formative feedback is provided during the presentation which should be considered for the final report.

The marking scheme for the report is based on the following criteria. Refer to the Assignment Brief for the assessment rubric for each criterion.

•    Problem analysis

•    Current solutions and potential technologies

•     Solution design

•    Professional and ethical issues

•    Critical evaluation

•    Ethical implications

•    Report presentation