Explainable Machine Learning for Real-Time Hypoglycemia and Hyperglycemia Prediction and Personalized Control Recommendations.

BACKGROUND: The occurrences of acute complications arising from hypoglycemia and hyperglycemia peak as young adults with type 1 diabetes (T1D) take control of their own care. Continuous glucose monitoring (CGM) devices provide real-time glucose readings enabling users to manage their control proactively. Machine learning algorithms can use CGM data to make ahead-of-time risk predictions and provide insight into an individual's longer term control. METHODS: We introduce explainable machine learning to make predictions of hypoglycemia (<70 mg/dL) and hyperglycemia (>270 mg/dL) up to 60 minutes ahead of time. We train our models using CGM data from 153 people living with T1D in the CITY (CGM Intervention in Teens and Young Adults With Type 1 Diabetes)survey totaling more than 28 000 days of usage, which we summarize into (short-term, medium-term, and long-term) glucose control features along with demographic information. We use machine learning explanations (SHAP [SHapley Additive exPlanations]) to identify which features have been most important in predicting risk per user. RESULTS: Machine learning models (XGBoost) show excellent performance at predicting hypoglycemia (area under the receiver operating curve [AUROC]: 0.998, average precision: 0.953) and hyperglycemia (AUROC: 0.989, average precision: 0.931) in comparison with a baseline heuristic and logistic regression model. CONCLUSIONS: Maximizing model performance for glucose risk prediction and management is crucial to reduce the burden of alarm fatigue on CGM users. Machine learning enables more precise and timely predictions in comparison with baseline models. SHAP helps identify what about a CGM user's glucose control has led to predictions of risk which can be used to reduce their long-term risk of complications.

Turning to Peers: Integrating Understanding of the Self, the Condition, and Others' Experiences in Making Sense of Complex Chronic Conditions.

People are increasingly involved in the self-management of their own health, including chronic conditions. With technology advances, the choice of self-management practices, tools, and technologies has never been greater. The studies reported here investigated the information seeking practices of two different chronic health populations in their quest to manage their health conditions. Migraine and diabetes patients and clinicians in the UK and the US were interviewed about their information needs and practices, and representative online communities were explored to inform a qualitative study. We found that people with either chronic condition require personally relevant information and use a broad and varied set of practices and tools to make sense of their specific symptoms, triggers, and treatments. Participants sought out different types of information from varied sources about themselves, their medical condition, and their peers' experiences of the same chronic condition. People with diabetes and migraine expended great effort to validate their personal experiences of their condition and determine whether these experiences were 'normal'. Based on these findings, we discuss the need for future personal health technologies that support people in engaging in meaningful and personalised data collection, information seeking, and information sharing with peers in flexible ways that enable them to better understand their own condition.

Can standards and regulations keep up with health technology?

Technology is changing at a rapid rate, opening up new possibilities within the health care domain. Advances such as open source hardware, personal medical devices, and mobile phone apps are creating opportunities for custom-made medical devices and personalized care. However, they also introduce new challenges in balancing the need for regulation (ensuring safety and performance) with the need to innovate flexibly and efficiently. Compared with the emergence of new technologies, health technology design standards and regulations evolve slowly, and therefore, it can be difficult to apply these standards to the latest developments. For example, current regulations may not be suitable for approaches involving open source hardware, an increasingly popular way to create medical devices in the maker community. Medical device standards may not be flexible enough when evaluating the usability of mobile medical devices that can be used in a multitude of different ways, outside of clinical settings. Similarly, while regulatory guidance has been updated to address the proliferation of health-related mobile phone apps, it can be hard to know if and when these regulations apply. In this viewpoint, we present three examples of novel medical technologies to illustrate the types of regulatory issues that arise in the current environment. We also suggest opportunities for support, such as advances in the way we review and monitor medical technologies.