Diabetes Technology Research Group, ARTORG Center for Biomedical Engineering Research, University of Bern, Bern, Switzerland.
Diabetes Technol Ther. 2012 Feb;14(2):168-74. doi: 10.1089/dia.2011.0093. Epub 2011 Oct 12.
Prediction of glycemic profile is an important task for both early recognition of hypoglycemia and enhancement of the control algorithms for optimization of insulin infusion rate. Adaptive models for glucose prediction and recognition of hypoglycemia based on statistical and artificial intelligence techniques are presented.
We compared an autoregressive (AR) model using only glucose information, an AR model with external insulin input (ARX), and an artificial neural network (ANN) using both glucose and insulin information. Online adaptive models were used to account for the intra- and inter-subject variability of the population with diabetes. The evaluation of the predictive ability included prediction horizons (PHs) of 30 min and 45 min.
The AR model presented root mean square error (RMSE) values of 14.0-21.6 mg/dL and correlation coefficients (CCs) of 0.92-0.95 for PH=30 min and 23.2-35.9 mg/dL and 0.79-0.87, respectively, for PH=45 min. The respective values for the ARX models were slightly better (PH=30 min, 13.3-18.8 mg/dL and 0.94-0.96; PH=45 min, 22.8-29.4 mg/dL and 0.83-0.88). For the ANN, the RMSE values ranged from 2.8 to 6.3 mg/dL, and the CC was 0.99 for all cases and PHs. The sensitivity of hypoglycemia prediction was 78% for AR, 81% for ARX, and 96% for ANN for PH=30 min and 65%, 67%, and 95%, respectively, for PH=45 min. The corresponding specificities were 96%, 96%, and 99% for PH=30 min and 93%, 93%, and 99% for PH=45 min.
The ANN appears to be more appropriate for the prediction of glucose profile based on glucose and insulin data.
血糖预测是识别低血糖和优化胰岛素输注率控制算法的重要任务。本研究提出了基于统计和人工智能技术的血糖预测和低血糖识别自适应模型。
我们比较了仅使用血糖信息的自回归（AR）模型、具有外源性胰岛素输入的 AR 模型（ARX）以及同时使用血糖和胰岛素信息的人工神经网络（ANN）。在线自适应模型用于考虑糖尿病患者的个体内和个体间变异性。预测能力的评估包括 30min 和 45min 的预测时程（PH）。
AR 模型在 PH=30min 时的均方根误差（RMSE）值为 14.0-21.6mg/dL，相关系数（CC）为 0.92-0.95；在 PH=45min 时，RMSE 值为 23.2-35.9mg/dL，CC 为 0.79-0.87。ARX 模型的相应值略好（PH=30min 时，13.3-18.8mg/dL，CC 为 0.94-0.96；PH=45min 时，22.8-29.4mg/dL，CC 为 0.83-0.88）。对于 ANN，RMSE 值范围为 2.8-6.3mg/dL，所有情况下的 CC 均为 0.99。在 PH=30min 时，AR、ARX 和 ANN 对低血糖的预测敏感性分别为 78%、81%和 96%，在 PH=45min 时，分别为 65%、67%和 95%。相应的特异性分别为 PH=30min 时的 96%、96%和 99%，PH=45min 时的 93%、93%和 99%。
ANN 似乎更适合基于血糖和胰岛素数据预测血糖谱。
