Mental illness diagnosis system based on cross-scale fusion method of neuroimaging

By using a cross-scale fusion method based on neuroimage based on cross-scale fusion method in the diagnosis of mental illness, voxel-level index information of brain multimodal magnetic resonance images is solved, and a higher diagnostic accuracy and information coverage are achieved.

CN119067922BActive Publication Date: 2025-03-25SHANDONG JIANZHU UNIV
2 Cites 0 Cited by

Patent Information

Application Number
CN202411079479.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-07
Publication Date
2025-03-25
Estimated Expiration
2044-08-07

AI Technical Summary

Technical Problem

The prior art ignores the covariant information and complementary information between different modal data in the diagnosis of mental illness, resulting in the inadequate mining of data information and the accuracy of the diagnostic model is reduced.

Method used

Using a cross-scale fusion method based on neural image, the multimodal magnetic resonance image of the brain is obtained and preprocessed, voxel-level index values ​​are extracted to form an index matrix, standardized and discrete Fourier transform fusion, frequency domain information and features are extracted, and the deep learning model is finally used to predict mental illness types.

Benefits of technology

It effectively explores the potential information inside and between the modals, improves the accuracy of model diagnosis, and provides more detailed and comprehensive information on brain functional connection abnormalities, helping doctors make more accurate diagnosis.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The present invention belongs to the field of brain imaging and deep learning, and provides a method and system for diagnosing mental illness based on a cross-scale fusion method of neural imaging. The method for diagnosing mental illness based on a cross-scale fusion method of neural imaging includes obtaining a multimodal magnetic resonance image of the brain of a subject and performing image preprocessing to obtain a multimodal magnetic resonance image of the brain with a uniform voxel size; extracting the index value of each voxel of the subject from the multimodal magnetic resonance image of the brain with a uniform voxel size to form an index matrix of the subject; normalizing each row of the index in the index matrix of the subject separately to obtain a standardized index matrix; using a discrete Fourier transform algorithm to fuse the data of each column in the standardized index matrix to obtain frequency domain information, and then using frequency domain distribution indicators to extract the features of the corresponding voxels; using a trained diagnostic model to process the features of each voxel, predicting the probability of the type of mental illness and outputting the type of mental illness with the highest probability.
Need to check novelty before this filing date? Find Prior Art