Brain entropy monitoring method and system for patient with alzheimer's disease based on electroencephalogram signal analysis

Abstract

Disclosed in the present invention are a brain entropy monitoring method and system for a patient with Alzheimer's disease based on electroencephalogram signal analysis. The method comprises: acquiring an original electroencephalogram signal of a patient with Alzheimer's disease, and denoising the original electroencephalogram signal to obtain a denoised electroencephalogram signal; decomposing the denoised electroencephalogram signal to obtain a low-frequency electroencephalogram signal associated with Alzheimer's disease; using a filtering algorithm based on independent component analysis to process the low-frequency electroencephalogram signal, so as to obtain a pure electroencephalogram signal; obtaining, based on the pure electroencephalogram signal, a high-quality electroencephalogram signal; and calculating, based on the high-quality electroencephalogram signal, a brain entropy value. The present invention uses a multi-scale entropy algorithm to analyze the processed electroencephalogram signal and calculate the brain entropy value, thereby evaluating the complexity of the brain across different time scales. This method, by means of multi-level signal processing and feature extraction, provides an important basis for early detection and intervention of Alzheimer's disease.

Images