Artificial Intelligence for Medical Systems (AIMS) Lab, Department of Biomedical Engineering, Oregon Health & Science University, Portland, Oregon, USA.
Harold Schnitzer Diabetes Health Center, Oregon Health and Science University, Portland, Oregon, USA.
Diabetes Technol Ther. 2020 Nov;22(11):801-811. doi: 10.1089/dia.2019.0458. Epub 2020 May 14.
Despite new glucose sensing technologies, nocturnal hypoglycemia is still a problem for people with type 1 diabetes (T1D) as symptoms and sensor alarms may not be detected while sleeping. Accurately predicting nocturnal hypoglycemia before sleep may help minimize nighttime hypoglycemia. A support vector regression (SVR) model was trained to predict, before bedtime, the overnight minimum glucose and overnight nocturnal hypoglycemia for people with T1D. The algorithm was trained on continuous glucose measurements and insulin data collected from 124 people (22,804 valid nights of data) with T1D. The minimum glucose threshold for announcing nocturnal hypoglycemia risk was derived by applying a decision theoretic criterion to maximize expected net benefit. Accuracy was evaluated on a validation set from 10 people with T1D during a 4-week trial under free-living sensor-augmented insulin-pump therapy. The primary outcome measures were sensitivity and specificity of prediction, the correlation between predicted and actual minimum nocturnal glucose, and root-mean-square error. The impact of using the algorithm to prevent nocturnal hypoglycemia is shown in-silico. The algorithm predicted 94.1% of nocturnal hypoglycemia events (<3.9 mmol/L, 95% confidence interval [CI], 71.3-99.9) with an area under the receiver operating characteristic curve of 0.86 (95% CI, 0.75-0.98). Correlation between actual and predicted minimum glucose was high ( = 0.71, < 0.001). In-silico simulations showed that the algorithm could reduce nocturnal hypoglycemia by 77.0% ( = 0.006) without impacting time in target range (3.9-10 mmol/L). An SVR model trained on a big data set and optimized using decision theoretic criterion can accurately predict at bedtime if overnight nocturnal hypoglycemia will occur and may help reduce nocturnal hypoglycemia.
尽管有新的葡萄糖感测技术，但 1 型糖尿病（T1D）患者仍存在夜间低血糖问题，因为在睡眠时可能无法检测到症状和传感器警报。在睡前准确预测夜间低血糖可能有助于最大程度地减少夜间低血糖。我们训练了一个支持向量回归（SVR）模型，以便在睡前预测 T1D 患者的 overnight minimum glucose 和 overnight nocturnal hypoglycemia。该算法使用来自 124 名（22804 个有效夜间数据）T1D 患者的连续葡萄糖测量值和胰岛素数据进行训练。夜间低血糖风险的最低血糖阈值是通过应用决策理论标准来最大化预期净收益得出的。在 10 名 T1D 患者进行为期 4 周的自由生活传感器增强胰岛素泵治疗试验的验证集中评估了准确性。主要的评估指标是预测的敏感性和特异性、预测的最低夜间血糖与实际最低夜间血糖之间的相关性，以及均方根误差。该算法对预防夜间低血糖的影响是在模拟环境中显示的。该算法预测了 94.1%的夜间低血糖事件（<3.9mmol/L，95%置信区间[CI]，71.3-99.9），其接受者操作特征曲线下面积为 0.86（95%CI，0.75-0.98）。实际和预测的最低血糖之间的相关性很高（ = 0.71， < 0.001）。模拟模拟表明，该算法可以在不影响目标范围内（3.9-10mmol/L）时间的情况下，将夜间低血糖减少 77.0%（ = 0.006）。使用大数据集训练并使用决策理论标准优化的 SVR 模型可以准确预测夜间是否会发生夜间低血糖，并可能有助于减少夜间低血糖。
