The global rise in type 2 diabetes underscores the need for scalable and cost-effective screening methods. Current diagnosis requires biochemical assays, which are invasive and costly. Advances in consumer wearables have enabled early explorations of machine learning-based disease detection, but prior studies were limited to controlled settings. We present SweetDeep, a compact neural network trained on physiological and demographic data from 285 (diabetic and non-diabetic) participants in the EU and MENA regions, collected using Samsung Galaxy Watch 7 devices in free-living conditions over six days. Each participant contributed multiple 2-minute sensor recordings per day, totaling approximately 20 recordings per individual. Despite comprising fewer than 3,000 parameters, SweetDeep achieves 82.5% patient-level accuracy (82.1% macro-F1, 79.7% sensitivity, 84.6% specificity) under three-fold cross-validation, with an expected calibration error of 5.5%. Allowing the model to abstain on less than 10% of low-confidence patient predictions yields an accuracy of 84.5% on the remaining patients. These findings demonstrate that combining engineered features with lightweight architectures can support accurate, rapid, and generalizable detection of type 2 diabetes in real-world wearable settings.

翻译：全球2型糖尿病患病率的上升凸显了对可扩展且经济高效的筛查方法的需求。目前的诊断依赖于生化检测，这些方法具有侵入性且成本高昂。消费级可穿戴设备的进步使得基于机器学习的疾病检测早期探索成为可能，但先前的研究仅限于受控环境。我们提出了SweetDeep，这是一种紧凑的神经网络，基于来自欧盟和中东及北非地区285名（糖尿病患者与非糖尿病患者）参与者的生理和人口统计数据训练而成，这些数据是在自由生活条件下使用三星Galaxy Watch 7设备在六天内收集的。每位参与者每天提供多个2分钟的传感器记录，总计每人约20条记录。尽管参数少于3,000个，SweetDeep在三折交叉验证下实现了82.5%的患者级别准确率（82.1%宏F1分数，79.7%灵敏度，84.6%特异性），预期校准误差为5.5%。允许模型对少于10%的低置信度患者预测进行弃权后，在剩余患者中准确率达到84.5%。这些发现表明，将工程化特征与轻量级架构相结合，可以在现实世界可穿戴设备环境中支持准确、快速且可泛化的2型糖尿病检测。