Method for classifying resting state functional magnetic resonance image data

• Main Authors: GUO HAO, LI YAO, YANG YANLI, LIU TAO
• Format: Patent
• Language: Chinese; English
• Published: 05.08.2022
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; IMAGE DATA PROCESSING OR GENERATION, IN GENERAL; PHYSICS

The invention relates to an image processing technology, in particular to a resting-state functional magnetic resonance image data classification method, which is based on a tree structure group lasso hypergraph U-Net model, and is realized by adopting the following steps: S1, preprocessing a resting-state functional magnetic resonance image; s2, carrying out region segmentation on the preprocessed resting state functional magnetic resonance image according to the selected standardized brain map, and carrying out average time sequence extraction on each segmented brain region; s3, calculating the correlation degree between every two average time sequences of each brain region by adopting a Pearson's correlation method so as to obtain a correlation matrix, and selecting a correlation coefficient value of a triangle on the matrix as a brain network feature of each subject; the method is suitable for magnetic resonance image data classification. 本发明涉及图像处理技术，具体是一种静息态功能磁共振影像数据分类方法，基于树结构组套索超图U-Net模型，该方法是采用如下步骤实现的：步