Alzheimer's disease detection method based on network graph theory analysis

Abstract

The invention discloses an Alzheimer's disease detection method based on network graph theory analysis. The method comprises the following steps: collecting data and preprocessing the data; calculating a voxel correlation matrix, calculating a multi-scale direction entropy of the voxel correlation matrix, obtaining a multi-scale direction entropy matrix, carrying out Z scoring, obtaining a score value matrix, carrying out weighted summation, obtaining a z score value map, and carrying out Gaussian kernel smoothing on the z score value map; segmenting and calculating the subcortical area volume and the total intracranial volume of the cerebral hemispheres on the two sides of the subject; and inputting the image after Gaussian kernel smoothing, the volume of the subcortical area and the total intracranial volume, selecting gene polymorphism sites, carrying out regression and statistical analysis, and if an output result is smaller than a preset value, judging that an Alzheimer's disease standard is met. The method is based on detection of gene polymorphic sites and magnetic resonance three-dimensional disturbed phase gradient echo sequence expansion, and can accurately evaluate the change of the subcortical nuclei volume under gene change.

Description

technical field [0001] The invention belongs to the technical field of medical detection, in particular to an Alzheimer's disease detection method based on network graph theory analysis. Background technique [0002] At present, the detection methods that use magnetic resonance to diagnose Alzheimer's disease include structural magnetic resonance, magnetic resonance spectroscopy, and magnetic resonance diffusion tensor imaging. The current detection methods all use network functional connectivity analysis methods based on independent component analysis or seed point pathways, which cannot fully describe the whole brain functional network connectome of Alzheimer's disease. At the same time, in previous studies on the structure of Alzheimer's disease, most of the applied detection techniques could not evaluate the deep nuclei structure. Contents of the invention [0003] In view of this, the present invention proposes a detection method for Alzheimer's disease based on the ...

AI Technical Summary

Problems solved by technology

The current detection methods all use the network functional connectivity analysis method based on independent component analysis or seed point approach, which cannot fully describe the whole brain functional network connectome of Alzheimer's disease.

At the same time, in previous Alzheimer's disease structural studies, most of the applied detection techniques could not evaluate the deep nuclei structure

Images