Non-invasive blood glucose detection method, processor and device based on graph convolution network

Abstract

The invention relates to the field of signal processing, and discloses a non-invasive blood sugar detection method, processor, device and storage medium based on a graph convolution network. The method includes: acquiring the PPG signal to be predicted; filtering the PPG signal to be predicted; converting the filtered PPG signal to be predicted into a node graph; obtaining a corresponding adjacency matrix and a feature matrix according to the node graph; inputting the adjacency matrix and the feature matrix into the graph Convolutional network, the corresponding blood sugar value is obtained through the graph convolutional network. Through the above technical solution, the present invention provides a deep learning method using a graph convolution network, which can make the model automatically find out the important feature information required for the blood sugar prediction problem, and at the same time, through the iterative update of the node information, In the case of retaining all feature information, the feature information is continuously optimized, so that the accuracy of blood glucose prediction is greatly improved.

Description

technical field [0001] The present invention relates to the field of signal processing, in particular, to a non-invasive blood glucose detection method, processor, device and storage medium based on graph convolution network. Background technique [0002] As a chronic disease, diabetes causes one death every seven seconds worldwide. There are no obvious symptoms in the early stage of diabetes, and when the complications become obvious, it often causes damage to other organs and causes complications. In order to monitor blood glucose concentration in real time and continuously, a variety of non-invasive blood glucose detection technologies have been developed rapidly in recent years, especially the application of photoplethysmography (PPG) technology to monitor blood glucose concentration. [0003] However, in the current state of the art, more research is to use traditional machine learning methods to build models, such as random forests, support vector machines, partial le...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to provide a non-invasive blood glucose detection method based on graph convolution network in order to overcome the problem of low accuracy of blood glucose prediction results in the prior art

Images