Robust Data-Driven Control of Artificial Pancreas Systems Using Neural Networks
Robust Data-Driven Control of Artificial Pancreas Systems Using Neural Networks
复制标题
DOI：
10.1007/978-3-319-99429-1_11
复制
发表时间：
2018-09
影响因子：
13.5
通讯作者：
Souradeep Dutta;Taisa Kushner;S. Sankaranarayanan
中科院分区：
文献类型：
--
作者：
Souradeep Dutta;Taisa Kushner;S. Sankaranarayanan
在本文中，我们提供了一种人工胰腺系统的数据驱动控制方法，通过从现有的患者数据中学习人类胰岛素-葡萄糖生理的神经网络模型，并使用混合整数优化方法来使用推断的模型实时控制血糖水平。首先，我们的方法学习神经网络来根据给定的胰岛素注射数据及其对血糖水平的影响来预测未来的血糖值。然而，为了对结果模型提供保证，我们使用分位数回归来拟合多个神经网络，这些神经网络除了预测平均值之外，还预测未来血糖水平的上下限分位数。使用推断的神经网络集，我们制定了一个模型预测控制方案，该方案同时调整基础和团注胰岛素的供应，以确保使用分位数模型将有害低血糖和高血糖的风险限制在有限的范围内，同时平均预测保持在尽可能接近期望的目标。我们讨论了这个方案如何处理来自大量未宣布膳食的干扰，以及由于未来血糖预测的不确定性太高而导致的不可行。我们根据从17名患者那里获得的数据对这种方法进行了实验评估，每个患者的疗程为40个晚上。此外，我们还使用从UVA-Padova 1型糖尿病模拟器获得的虚拟患者模型获得的神经网络来测试我们的方法。
In this paper, we provide an approach to data-driven control for artificial pancreas systems by learning neural network models of human insulin-glucose physiology from available patient data and using a mixed integer optimization approach to control blood glucose levels in real-time using the inferred models. First, our approach learns neural networks to predict the future blood glucose values from given data on insulin infusion and their resulting effects on blood glucose levels. However, to provide guarantees on the resulting model, we use quantile regression to fit multiple neural networks that predict upper and lower quantiles of the future blood glucose levels, in addition to the mean.Using the inferred set of neural networks, we formulate a model-predictive control scheme that adjusts both basal and bolus insulin delivery to ensure that the risk of harmful hypoglycemia and hyperglycemia are bounded using the quantile models while the mean prediction stays as close as possible to the desired target. We discuss how this scheme can handle disturbances from large unannounced meals as well as infeasibilities that result from situations where the uncertainties in future glucose predictions are too high. We experimentally evaluate this approach on data obtained from a set of 17 patients over a course of 40 nights per patient. Furthermore, we also test our approach using neural networks obtained from virtual patient models available through the UVA-Padova simulator for type-1 diabetes.