ASPECT METHOD FOR ESTIMATING ASPECT SCORE WITH THE PROCESS OF SEPARATING LESIONS RELATED TO STROKE

摘要

A method of estimating an ASPECT score based on a brain CT image of a stroke patient according to the present invention includes: a preprocessing step in which the preprocessing unit normalizes and standardizes the features of the image dataset; A segmentation step in which the image processing unit separates each lesion within a CT image classified into a supra ganglion level and a ganglion level; And a judging unit independently constructing a neural network for learning positive/negative images for each lesion to determine whether the lesion is stroked, including, but, wherein the segmentation step extracts seven regions at the ganglion level from the brain CT image. Then, extracting 3 regions from the Supra ganglion level region and extracting a total of 10 regions, and using the skull ellipse information obtained in the pre-processing step and feature points included in the extracted 10 regions, each is based on image processing technology. It includes the step of performing the segmentation on the lesion, and after searching for the feature points based on the preset division criteria, each feature point can be geometrically connected to divide the lesion.