Self-managing chronic conditions typically involves a diverse network of individuals and devices, forming a self-management ecosystem. For this ecosystem to be effective, components need to work together cohesively. The rapid advancement of technology means new devices need to be repeatedly integrated into existing self-management ecosystems. To examine this process, we used the case study of young adults with type 1 diabetes (T1D) in the UK who were given a smartwatch. Over six months, interviews and focus groups were performed to explore their smartwatch use alongside T1D management. Thematic analysis highlighted that smartwatches have potential as a display, interface and data source in T1D management, which is of particular importance as artificial intelligence plays a growing role in self-management ecosystems. It also emphasised the need for customisation, flexibility and adaptability, and automation in the design of technology to promote integration into existing self-management ecosystems for both T1D and other chronic conditions.

Reinforcement learning (RL) holds promise for supporting personalised decision-making in healthcare, but existing approaches often struggle to incorporate patient expertise and individual preferences, key components for clinically viable AI systems. This work introduces PAINT (Preference Adaptation for Individualised Treatment), a general framework for preference-guided offline RL in safety-critical settings. PAINT combines sketch-based reward annotation with safety-constrained policy optimisation, enabling fine-grained preference capture from historical patient data without requiring action labels. A reward model trained on this feedback guides offline RL while supporting tunable sensitivity to preference signals and enforcing clinical safety constraints. Using type 1 diabetes (T1D) management as a case study, in-silico evaluation with the FDA-accepted T1D simulator demonstrates that can PAINT reduces patient risk by 15% over commercial baselines under guidance, while enabling preference-driven adaptations such as improved management during challenging mealtime events and enhanced robustness to dosing errors. The method further shows resilience to real-world challenges including sample size, annotation noise, and inter-patient variability. These findings suggest PAINT offers a practical pathway for integrating human feedback into offline RL in patient settings, with broader implications for developing trustworthy and adaptive AI systems in healthcare.

Engaging end user groups with machine learning (ML) models can help align the design of predictive systems with people's needs and expectations. We present a co-design study investigating the benefits and challenges of using computational notebooks to inform ML models with end user groups. We used a computational notebook to engage young adults, carers, and clinicians with an example ML model that predicted health risk in diabetes care. Through co-design workshops and retrospective interviews, we found that participants particularly valued using the interactive data visualisations of the computational notebook to scaffold multidisciplinary learning, anticipate benefits and harms of the example ML model, and create fictional feature importance plots to highlight care needs. Participants also reported challenges, from running code cells to managing information asymmetries and power imbalances. We discuss the potential of leveraging computational notebooks as interactive co-design tools to meet end user needs early in ML model lifecycles.

Self-managing a chronic condition involves adapting management strategies to life’s continual change. Among these changes, moments of significant life transition can render routine self-management practices obsolete without significant modification to the new context. In this study, we examine one significant life transition for young adults living with Type 1 Diabetes, the move from home to university, to understand how near future AI-enhanced technologies might provide opportunities and challenges for supporting care. From interviews with 24 students in the UK who had moved away from their childhood homes, we used sensemaking literature to frame the process of initial disruption to the rebuilding of self-care practices around a new lifestyle and support networks. By studying a significant life transition, we uncover implications for the design of T1D technology, particularly closed-loop systems, through AI enhancements and human-centred design approaches, then extrapolate for other significant life transitions and chronic conditions.