Brain image-based infarction region segmentation method and device, equipment and medium

Abstract

The invention provides an infarction region segmentation method and device based on a brain image, equipment and a medium. The method comprises the steps of determining a seed point region for infarction region segmentation based on a candidate ventricle region and a cerebral cortex region extracted from a brain image; determining an infarction segmentation point in the candidate ventricle region by using an optimal path from each pixel point in the seed point region to a central point in the brain image; and segmenting a corresponding infarction region from the candidate ventricle region by using the infarction segmentation point. According to the method, the ant colony algorithm is adopted to analyze the infarction segmentation points in the candidate ventricle region, so that the corresponding infarction region is segmented from the candidate ventricle region by using the infarction segmentation points, the infarction region adhered to cerebrospinal fluid in the candidate ventricle region is automatically and effectively segmented, and the universality and efficiency of infarction region segmentation are enhanced.

Description

technical field [0001] The embodiments of the present application relate to the technical field of image processing, and in particular to a brain image-based infarction region segmentation method, device, device, and medium. Background technique [0002] Acute ischemic cerebrovascular disease, also known as acute ischemic stroke, is a major disease that seriously threatens the health of our population and hinders social and economic development, with high disability and fatality rates. [0003] In the treatment of acute ischemic cerebrovascular disease, computerized tomography perfusion imaging (CTP) can usually be used to conduct continuous CT scans of the patient's brain at different levels, so as to obtain the corresponding brain video. Then, according to the atlas of various perfusion parameters represented in the scanned brain image, it is judged whether there is a cerebral infarction, and then the corresponding cerebral infarction area is determined. [0004] However...

AI Technical Summary

Problems solved by technology

[0004] However, since brain images belong to CT single-mode images, and the cerebrospinal fluid in the ventricle has the same performance as the infarct lesion on certain two perfusion parameters represented in the brain images, the gray scale of the cerebrospinal fluid and infarct area in the brain images The actual volume of the brain infarction in the brain image cannot be accurately identified due to the presence of adhesions in three-dimensional space

Images