Online fault detection method of continuous blood sugar monitoring sensor based on multi-model fusion

Abstract

The invention relates to an online fault detection method of a continuous blood sugar monitoring sensor based on multi-model fusion, which comprises the following steps of: S1, acquiring online CGM monitoring signal data; S2, inputting the acquired on-line CGM monitoring signal data into a multi-model fusion algorithm model to acquire an on-line prediction error; S3, calculating the obtained on-line prediction error and the historical prediction error in combination to obtain the entropy value of the on-line time; S4, comparing the entropy values Ji1 and Ji2 of the online time obtained by calculation with the threshold values Tkl1 and Tkl2 of the current time respectively; if the entropy values Ji1 and Ji2 at the current moment are not all larger than the threshold values Tkl1 and Tkl2 atthe current moment, judging that the current blood sugar monitoring sensor works normally; if the entropy values Ji1 and Ji2 at the current moment are both larger than the threshold values Tkl1 and Tkl2 at the current moment, judging that the current blood sugar monitoring sensor works abnormally. The detection method has the advantage of high detection precision.

Description

technical field [0001] The invention belongs to the technical field of blood sugar monitoring, in particular to an online fault detection method for continuous blood sugar monitoring sensors based on multi-model fusion. Background technique [0002] The artificial pancreas (AP) system provides automatic regulation of blood glucose concentration (BGC) for type 1 diabetes (T1D) patients, and it mainly consists of three parts: continuous glucose monitoring (CGM) sensor, control of insulin infusion rate calculated based on CGM signal controller, and an insulin pump that delivers the amount of insulin calculated by the controller to the patient. Patients with T1D can have a more comprehensive understanding of blood sugar fluctuations through continuous monitoring of blood sugar, so as to better control blood sugar. However, in real life, the measurement results of the continuous blood glucose monitoring sensor are affected by many factors, resulting in inaccurate measurement res...

AI Technical Summary

Problems solved by technology

[0005] In view of the existing technical problems, the present invention provides an online fault detection method for continuous blood glucose monitoring sensors based on multi-model fusion, which solves the problems of low accuracy of detection results in the prior art

Images