Magnetic resonance brain image based subject individual difference data relation analysis method

Abstract

The invention provides a magnetic resonance brain image based subject individual difference data relation analysis method. The method includes: A, acquiring a magnetic resonance brain image of each subject; B, establishing a multidimensional variable of each subject on the basis of at least one unit to be analyzed of the magnetic resonance brain image of each subject, and at least one parameter variable corresponding to each unit to be analyzed; C, calculating a characteristic distance between the multidimensional variables of every two subjects according to the multidimensional variable of each subject, and establishing a first characteristic distance matrix of each subject according to the characteristic distances; D, acquiring a second kind of data of each subject, calculating a characteristic distance between the second kind of data of every two subjects, and establishing a second characteristic distance matrix of each subject according to the characteristic distances; and E, determining the correlation between the parameter variable of the unit to be analyzed of each subject and the second kind of data according to the similarity between the first characteristic distance matrix and the second characteristic distance matrix.

Description

technical field [0001] The invention relates to the field of brain imaging, in particular to a data relationship analysis method based on the magnetic resonance brain images of the subjects and the individual differences among the subjects. Background technique [0002] At present, current image-based neuroscience analysis methods usually examine the relationship between the brain image information of a single analysis unit and other data other than the analyzed image information. For example, in the prior art, when comparing the brain image differences between congenital and normal people, for the complete brain image of each subject, it is assumed that it is composed of n brain regions, and only the brain images are analyzed independently one by one. For these n brain regions, the individual brain regions of the brain images of each subject were selected and compared and analyzed one by one, in an attempt to explore the differences between the brain regions of congenital s...

AI Technical Summary

Problems solved by technology

This method ignores the relative information between different analysis units of the image data, that is, only considers multiple brain regions independently, but ignores the relative information between each brain region among n brain regions, that is, does not consider each brain region. Variation and interrelationship between regions

[0003] At the same time, most of the current neuroimaging analysis methods for magnetic resonance imaging data use the comparison of differences between groups based on the original data of brain imaging, correlation analysis between subjects, etc., ignoring the difference between two subjects as the basic unit of data for analysis.

[0004] It can be seen that the existing technology ignores the relative information between the multiple analysis units of the subject's individual brain image and the individual difference information between the two subjects, which partially hinders the analysis of the subject's brain image information and the images that are analyzed. The study of relationships with data other than information

Images