Transdisciplinary Department of Medicine & Advanced Technology, Seoul National University Hospital, 101 Daehak-ro, Jongro-gu, Seoul 03080, Republic of Korea; Institute of Medical and Biological Engineering, Seoul National University, Seoul 03080, Republic of Korea.
Department of Biomedical Engineering, Seoul National University College of Medicine, 103 Daehak-ro, Jongro-gu, Seoul 03080, Republic of Korea; Institute of Bioengineering, Seoul National University, Gwanak-ro 1, Seoul 08826, Republic of Korea; Artificial Intelligence Institute, Seoul National University, Seoul, 08826, Republic of Korea.
Comput Methods Programs Biomed. 2023 Oct;240:107694. doi: 10.1016/j.cmpb.2023.107694. Epub 2023 Jun 29.
Complete identification of the glucose dynamics for a patient generally requires prior clinical procedures and several measurements for the patient. However, these steps may not be always feasible. To address this limitation, we propose a practical approach integrating learning-based model predictive control (MPC), adaptive basal and bolus injections, and suspension with minimal requirements of prior knowledge of the patient.
The glucose dynamic system matrices were periodically updated using only input values, without any pretrained models. The optimal insulin dose was calculated based on a learning-based MPC algorithm. Meal detection and estimation modules were also introduced. The basal and bolus insulin injections were fine-tuned using the performance of glucose control from the previous day. To validate the proposed method, evaluations with 20 virtual patients from a type 1 diabetes metabolic simulator were employed.
Time-in-range (TIR) and time-below-range (TBR) were 90.8% (84.1% - 95.6%) and 0.3% (0% - 0.8%), as represented by the median, first (Q1), and third quartiles (Q3), respectively, when meal intakes were fully announced. When one out of three meal intake announcements was missing, TIR and TBR were 85.2% (75.0% - 88.9%) and 0.9% (0.4% - 1.1%), respectively.
The proposed approach obviates the need for prior tests from patients and shows effective regulation of blood glucose levels. From the perspective of practical implementation in clinical environments, to deal with minimal prior information of the patient, our study demonstrates how essential clinical knowledge and learning-based modules can be integrated into a control framework for an artificial pancreas.
通常情况下，全面了解患者的血糖动态变化需要进行前期临床检查并对患者进行多次测量。然而，这些步骤并非总是可行。为了解决这一局限性，我们提出了一种实用的方法，该方法将基于学习的模型预测控制（MPC）、自适应基础率和推注量以及最小化患者先验知识要求的暂停相结合。
仅使用输入值周期性地更新血糖动态系统矩阵，而无需任何预训练模型。基于基于学习的 MPC 算法计算最佳胰岛素剂量。还引入了进餐检测和估计模块。使用前一天血糖控制的性能对基础率和推注量胰岛素进行微调。为了验证所提出的方法，我们使用来自 1 型糖尿病代谢模拟器的 20 个虚拟患者进行了评估。
当进餐摄入被完全告知时，时间在目标范围内（TIR）和时间低于目标范围（TBR）分别为 90.8%（84.1%-95.6%）和 0.3%（0%-0.8%），中位数、第一分位数（Q1）和第三分位数（Q3）分别表示。当三分之一的进餐摄入通告中缺少一个时，TIR 和 TBR 分别为 85.2%（75.0%-88.9%）和 0.9%（0.4%-1.1%）。
所提出的方法无需患者进行前期测试，并且可以有效地调节血糖水平。从在临床环境中实际实施的角度来看，为了处理患者的最小先验信息，我们的研究展示了临床知识和基于学习的模块如何整合到人工胰腺的控制框架中。
