Auxiliary diagnosis system and method for alzheimer disease based on dynamic brain network graph kernel

Abstract

The invention provides an auxiliary diagnosis system and method for the alzheimer disease based on the dynamic brain network graph kernel. The diagnosis system comprises a preprocessing unit, a dynamic brain network construction unit, a dynamic brain network graph kernel calculation unit and a classification diagnosis unit, firstly, image preprocessing is conducted on a functional magnetic resonance imaging image by the preprocessing unit, then brain region matching, time period division, mutual information value calculation and frequent subgraph mining are conducted on the preprocessed functional magnetic resonance imaging image in sequence by the dynamic brain network construction unit, and bipartite graph optimal matching, graph kernel calculation, graph kernel matrix combination operation and weight distribution are conducted on a reconstructed frequent subgraph dynamic brain network in sequence by the dynamic brain network graph kernel calculation unit; a fused dynamic brain function network graph kernel matrix is obtained, in combination with a kernel SVM, data training is conducted through a data trainer, and diagnosis for the alzheimer disease is finally achieved through anauxiliary diagnosis device.

Description

Technical field [0001] The present invention relates to the technical field of computer-assisted diagnosis, in particular to an Alzheimer's disease auxiliary diagnosis system and method based on dynamic brain network graph cores. Background technique [0002] Alzheimer's disease (Alzheimer Disease, AD) is a neurodegenerative disease that affects people's cognition and even behavior. The cause of this disease is that some brain areas that control cognition are damaged, leading to connections between brain areas Attenuation or disappearance, therefore, the brain connection pattern plays a key role in the diagnosis of AD. Resting state functional magnetic resonance imaging (rs-fMRI) provides a non-invasive method to measure the functional activity and changes of the brain. Taking brain regions as nodes and the connections between brain regions as edges, construct a brain network based on rs-fMRI, classify the brain network, and realize AD diagnosis. [0003] The traditional static b...

AI Technical Summary

Problems solved by technology

However, the existing graph kernels are all based on static brain networks and only consider the global functional connectivity between brain regions

Images