Automatic sketching method and sketching system for brain function region of MRI head image, computing device and storage medium

Abstract

The invention discloses an automatic sketching method and sketching system for a brain function region of an MRI head image, a computing device and a storage medium, and the method comprises the steps: obtaining a preset number of T1MRI brain images, and making 116 brain function region segmentation tags of each image; performing same preprocessing and 2.5 D block processing on the MRI brain image and the brain function region segmentation tag to obtain respective 2.5 D data blocks; inputting the 2.5 D data blocks into the established semantic segmentation neural network for training until the model is stably converged, and obtaining an optimal brain functional region segmentation neural network model; inputting the to-be-segmented image subjected to the same preprocessing and 2.5 D block processing into the trained brain function region segmentation neural network model to obtain a brain function region segmentation result; and performing post-processing and edge detection on the brain function region segmentation result to obtain a brain function region contour sketching result. The method can achieve the automatic segmentation of the human brain function region, improves the speed and accuracy of the segmentation of the brain function region, and also improves the robustness and adaptability of the segmentation of the brain function region.

Description

technical field [0001] The invention relates to the technical field of image recognition, in particular to a method for automatically delineating brain functional areas of MRI head images, a delineating system, computing equipment and storage media. Background technique [0002] MRI images are widely used in structural pathological and functional research and clinical applications of the brain. In the field of neurosurgery, preoperative MRI imaging is used to determine the location and size of lesions and surgical planning, relying on accurate segmentation of brain functional areas around lesions from MRI brain imaging. [0003] The current theory of MRI brain functional area segmentation is mainly based on the Anatomical Automatic Labeling (AAL) brain functional area template provided by the Montreal Neurological Institute. The AAL template has a total of 116 areas, of which 90 are belong to the cerebrum, and the remaining 26 belong to the cerebellar structure. [0004] A...

AI Technical Summary

Problems solved by technology

[0004] At present, the algorithm based on AAL template matching is usually used to segment brain functional regions in T1MRI images, but this method has high computational complexity and time-consuming, and has anti-noise and adaptability to brain images with incomplete scanning of brain images and large structural variations. Poor stability and other problems, in the actual clinical application, the stability and convenience are not high

Images