Multi-feature fusion data classification method based on brain function super network model

Abstract

The invention relates to an image processing technology, in particular to a multi-feature fusion magnetic resonance image data classification method based on a brain function super network model. According to the invention, the problem of low accuracy caused by the fact that a traditional magnetic resonance imaging data classification method uses a single attribute feature is solved, the topological structure of the super-network is evaluated in a multi-angle three-dimensional mode, the integrity of topological information of the super-network is presented, the inter-group difference characterization capability is enhanced, and the method is suitable for related research of functional magnetic resonance image classification. According to the invention, firstly, a complex MCP method is usedfor constructing a super-network model, and then 11 different features are extracted from the super-network to serve as fusion features for classification, so that the defect that a single attributefeature contains single information is overcome. The feature set rich in sufficient information can represent omnibearing multi-angle topology in the brain super-network, and the integrity of a super-network topology structure is presented, so that it is ensured that a classifier constructed later can effectively extract discrimination information, and the upper limit of the classification precision of the classifier is improved.

Description

technical field [0001] The invention relates to image processing technology, in particular to a multi-feature fusion data classification method based on a brain function hypernetwork model. Background technique [0002] In recent years, neuroimaging techniques have been increasingly used to study brain-region interactions. There is a significant low-frequency correlation in the low-frequency blood oxygenation level dependent (BOLD) signal, which can be used as a neurophysiological indicator to detect spontaneous brain activity in a resting state. We can construct functional brain networks through BOLD signaling. Functional brain networks have been widely used in the field of neuropsychiatric diseases, helping to clarify the pathological mechanisms of brain and psychiatric diseases, and may provide related imaging markers, providing a new perspective for the diagnosis and evaluation of clinical brain diseases. [0003] Based on the image data obtained by functional magnetic...

AI Technical Summary

Problems solved by technology

[0006] In order to solve the problem of one-sided information and low classification accuracy in the single-attribute feature quantification brain function super-network model, the present invention provides a functional magnetic resonance image data classification method based on super-network fusion features

Images