Using machine learning methods to predict the risk of exercise-induced hypoglycemia in people with Type 1 diabetes

Exercise offers a number of benefits to individuals with type 1 diabetes (T1D) (1), including reduced risk of coronary artery disease and stroke, as well as improved blood pressure (2,3), lowered daily insulin requirements (3), and reduced diabetic complications, such as retinopathy and neuropathy (4).
The American Diabetic Association recommends 150 minutes of aerobic exercise a week for people with T1D (5) but few are reaching this target (3,6,7). One of the main reasons for this is fear of exercise-induced hypoglycaemia (8–11).
Hypoglycemia occurs when blood glucose (BG) drops below 3.9 mmol/L and can vary in severity, with side effects ranging from dizziness, anxiety and nausea to unconsciousness, seizures and even death (12). Exercised-induced hypoglycaemia can occur not only during exercise, but for up to 36 after exercise and also overnight (6,13).
In addition, understanding glycaemic response to exercise is challenging for people with T1D since it is affected by so many factors. Firstly, different types of exercise will have different effects on BG levels, for example aerobic exercise will decrease BG levels whereas anaerobic exercise will increase levels (1). The glycemic response is also affected by duration and intensity of exercise, amount of insulin on board, location of insulin delivery, the starting BG before exercise, and the last meal or snack (14). This is further complicated by the fact that people with T1D often will not notice their symptoms when exercising (15). This complex interaction of factors make hypoglycaemia around exercise very difficult to predict and prevent.
It is therefore understandable that people with T1D are less active and reluctant to exercise, when in addition to common barriers to exercise reported in the general population, they may also experience the discomfort of changes in glucose levels and heightened risk of hypoglycaemia.
Wearable sensors are becoming increasingly common tools in the management of T1D. From continuous glucose monitors (CGMs) and insulin pumps, to smartphone apps, heart rate (HR) monitors, and physical activity trackers available (16), the accessibility of these devices is opening up many opportunities to get relevant data. If we can harness data from these devices to effectively predict hypoglycaemia around exercise, we can offer people with T1D practical guidance for effectively avoiding hypoglycaemia around exercise, enabling them to feel more confident about exercising.
Our goal is to utilise wearable technology paired with machine learning techniques to develop a tool that predicts whether hypoglycaemia is going to occur during or after exercise and provide them with simple, actionable guidance. With this, we hope to assist people with T1D allow patients to take action to avoid hypoglycaemia and thus exercise more safely.

运动对1型糖尿病（Type 1 Diabetes, T1D）患者具有诸多获益[1]，包括降低冠状动脉疾病与中风风险、改善血压水平[2,3]、减少每日胰岛素需求量[3]，以及缓解糖尿病并发症（如视网膜病变与神经病变）[4]。
美国糖尿病协会（American Diabetic Association）建议1型糖尿病患者每周完成150分钟的有氧运动[5]，但鲜有患者能够达到这一目标[3,6,7]，其中一项主要原因是对运动诱发性低血糖（exercise-induced hypoglycaemia）的恐惧[8–11]。
低血糖指血糖（Blood Glucose, BG）降至3.9 mmol/L以下，其严重程度各不相同，不良反应涵盖头晕、焦虑、恶心，乃至意识丧失、惊厥甚至死亡[12]。运动诱发性低血糖不仅可发生于运动过程中，还可在运动后长达36小时乃至夜间出现[6,13]。
此外，由于受诸多因素影响，1型糖尿病患者难以准确把握运动后的血糖应答。其一，不同运动类型对血糖水平的影响存在差异：例如有氧运动可降低血糖，而无氧运动则会升高血糖[1]。血糖应答还受运动时长、强度、体内存留胰岛素量、胰岛素注射部位、运动前初始血糖水平以及末次进餐或加餐等因素影响[14]。更为复杂的是，1型糖尿病患者在运动时常无法察觉自身的低血糖症状[15]。这些复杂的交互因素使得运动前后的低血糖难以预测与预防。
因此，相较于普通人群常见的运动障碍，1型糖尿病患者还需承受血糖波动带来的不适，以及更高的低血糖风险，故而他们活动量更少、对运动心存顾虑也就不难理解了。
可穿戴传感器正逐渐成为1型糖尿病管理的常用工具。从连续血糖监测仪（Continuous Glucose Monitors, CGMs）、胰岛素泵，到智能手机应用程序、心率（Heart Rate, HR）监测仪与运动追踪设备[16]，这类设备的普及为获取相关临床数据提供了诸多契机。若能利用这些设备采集的数据，有效预测运动前后的低血糖风险，便可为1型糖尿病患者提供切实可行的指导，帮助其有效规避运动相关低血糖，从而更有信心地参与运动。
本研究旨在将可穿戴技术与机器学习技术相结合，开发一款可预测运动过程中或运动后是否会发生低血糖的工具，并为患者提供简单易懂、可执行的指导。借此，我们期望助力1型糖尿病患者采取措施规避低血糖，进而更安全地开展运动。