Co-designing opportunities for Human-Centred Machine Learning in supporting Type 1 diabetes decision-making

Aug 3, 2026

Upload by Jiven

Type 1 diabetes self-management involves hundreds of daily decisions, and machine-learning-enabled diabetes technologies can simplify care while improving decision support. Yet they often depend on burdensome data logging and cognitively demanding reflection on collected information. This work uses co-design to find practical opportunities for machine learning support in everyday settings. After nine months of interviews and design workshops with 15 participants, assumptions about user needs were revised: users trust their own knowledge during routine situations and reject ML decision support, but they request guidance in unfamiliar or unexpected contexts. The paper discusses how ML and other AI approaches, such as expert systems, could support decisions both routinely and under uncertainty.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages19

Reading Time48 min

AccessFree to download

Introduction

Diabetes management challenges

Limitations of existing diabetes technologies

Machine learning decision support opportunities

| Co-designing opportunities for Human-Centred Machine Learning in supporting Type 1 diabetes decision-making
Katarzyna Stawarz a,∗, Dmitri Katz b, Amid Ayobi c, Paul Marshall d, Taku Yamagata d, Raul Santos-Rodriguez d, Peter Flachd, Aisling Ann O’Kane d
a School of Computer Science and Informatics, Cardiff University, Abacws, Senghennydd Road, CF24 4AG Cardiff, United Kingdom
b The Open University, Walton Hall, MK7 6AA Milton Keynes, United Kingdom
c UCL Interaction Centre, University College London, 66-72 Gower Street, London, WC1E 6EA, United Kingdom d Department of Computer Science, University of Bristol, Queen’s Building, University Walk, BS8 1TR Bristol, United Kingdom | | | | | --- | --- | --- | --- | | A R T I C L E I N F O | | A B S T R A C T | | | Keywords: Diabetes Health
Qualitative research Machine learning Decision support Artificial intelligence | | Type 1 Diabetes (T1D) self-management requires hundreds of daily decisions. Diabetes technologies that use machine learning have significant potential to simplify this process and provide better decision support, but often rely on cumbersome data logging and cognitively demanding reflection on collected data. We set out to use co-design to identify opportunities for machine learning to support diabetes self-management in everyday settings. However, over nine months of interviews and design workshops with 15 people with T1D, we had to re-assess our assumptions about user needs. Our participants reported confidence in their personal knowledge and rejected machine learning based decision support when coping with routine situations, but highlighted the need for technological support in the context of unfamiliar or unexpected situations (holidays, illness, etc.). However, these are the situations where prior data are often lacking and drawing data-driven conclusions is challenging. Reflecting this challenge, we provide suggestions on how machine learning and other artificial intelligence approaches, e.g., expert systems, could enable decision-making support in both routine and unexpected situations. | | 1. Introduction Among major health conditions, diabetes is one of the most common, affecting over 400 million people worldwide (International Diabetes Federation, 2017). Type 1 diabetes (T1D), which affects 5%–10% of those with diabetes, is an autoimmune condition requiring frequent injections of insulin to maintain blood glucose (BG) levels within a safe range. Elevated levels (hyperglycaemia) can lead to long-term complications, such as blindness, kidney failure, or nerve damage, while severely low BG levels (hypoglycaemia) can lead to unconsciousness, seizure, coma, and – in rare cases – death (McGill and Ahmann, 2017). While clinicians can play an important role in supporting diabetes care, effective daily management relies primarily on an individual’s habits and management decisions (Funnell and Anderson, 2004). Selfmanaging diabetes typically involves self-monitoring BG levels and lifestyle factors such as food and physical activity multiple times per day, analysing this information, and dynamically adjusting numerous factors accordingly (Klonoff, 2012). However, maintaining this balance with the demands of daily life is challenging, resulting in many individuals failing to meet clinical guidelines (Miller et al., 2015). It is ∗ Corresponding author. E-mail address: [stawarzk@cardiff.ac.uk](stawarzk@cardiff.ac.uk) (K. Stawarz). therefore important to find new approaches that will help individuals with diabetes make better informed decisions while reducing the burden of care. Currently, people with diabetes have access to several types of technologies, including continuous glucose meters (CGM), insulin pumps, smartphone apps and, more recently, hybrid-closed loop systems which integrate these systems (National Institute for Health and Excellence, 2023). However, with the complex and personal nature of the condition (Mol, 2008; O’Kane

FAQs

How does the paper propose to address decision support under routine and unexpected conditions?

It suggests using machine learning and other AI methods, including expert systems, to enable actionable guidance for both well-known and uncertain situations.

Aug 3, 2026

Upload by Jiven