Fine-Scale Online Evaluation of Glycemic Control Performance Based on Temporal Feature Analysis

Abstract

Traditional glycemic control metrics ignore the time axis of continuous glucose monitoring (CGM) data which thus cannot further elaborate control performance of glucose within the normal region. They cannot be implemented in real time either, which may lead to incomplete assessment. A fine-scale assessment method of glycemic control performance is proposed for glucose within the normal range. The glucose control performance within the normal glycemic range is divided into three categories including accurate control, benign control, and suboptimal control by analyzing the temporal changing speed of the CGM data. First, the temporal slowness is extracted based on slow feature analysis (SFA) to capture the temporal variations of CGM data considering the time component. Then a monitoring statistic is developed for the temporal slowness. Finally, the control assessment grid analysis is proposed for both qualitative analysis and quantitative analysis. The proposed method is investigated using data collected from 10 clinical adults with type 1 diabetes. The results show that the proposed method can evaluate the control performance in real time. The proposed metric can be regarded as a good auxiliary metric to improve the control performance assessment for glucose within the normal region, in which the percentages of points falling in accurate control zone A is about 43.80% for the 10 subjects. It is the first time that the concept of online fine-scale assessment for glycemic control performance is proposed which extracts the temporal slow features by quantitative analysis of the temporal distribution of CGM data.

Semantic Knowledge Integrated Graph Convolutional Network for Zero-Sample Tracing of Control Performance Degradation

“…Huang proposed a probabilistic framework to synthesize various CPD detection techniques based on Bayesian methods. As for controller problems, the minimum variance control benchmark, , LQG benchmark, , and other benchmarks − are recommended. Some methods − are employed to detect the stiction problems of control valves.…”

An Enhanced Method to Assess MPC Performance Based on Multi-Step Slow Feature Analysis

“…This method was successfully applied to batch processes monitoring [35], as well as soft sensor modeling [36]. Zhao et al [37] proposed a fine-scale assessment method of glycemic control performance by analyzing the temporal changing speed of the monitored data. By analyzing the process dynamics directly related to closed-loop control from quality-relevant view for industrial processes, Qin and Zhao [38] proposed a fine-scale monitoring method of process performance status.…”