Method for calculating ASPECTS score of non-enhanced CT

Abstract

The invention provides a method for calculating an ASPECTS score of a non-enhanced CT, and the method comprises the following steps: obtaining cerebral ischemia non-enhanced CT image data, and extracting a target region; judging whether ischemia occurs or not, and marking; designing a cerebral ischemia classification model based on deep learning, and training the classification model by using the marked cerebral ischemia non-enhanced CT image data, the input data are an image to be detected, mask images of corresponding areas of the left and right brains and a classification label for judging whether ischemia exists or not; and calculating an ASPECTS score according to the obtained classification result. The method not only considers the local comparison of the left and right sides of the same region, but also considers the global situation of the region in the whole brain tissue.

Description

technical field [0001] The invention relates to the fields of medical images and computers, in particular to an ASPECTS scoring method for calculating non-enhanced CT. Background technique [0002] Alberta Stroke Program Early CT Score (ASPECTS) is a simple, reliable and systematic method to evaluate early ischemic changes in the middle cerebral artery supply area in patients with ischemic stroke. Mainly based on non-enhanced CT scan (NCCT), 10 regions in two layers of the middle cerebral artery blood supply area were selected on the CT image, such as figure 1 Shown are 10 areas on one side, and the opposite side is completely symmetrical to the area shown: ①The level of nuclei (that is, the plane of the thalamus and striatum), divided into M1, M2, M3, insula I, lenticular nucleus L, There are 7 areas of caudate nucleus C and hindlimb IC of the internal capsule; ②The level above the nucleus (2cm above the level of the nucleus), including M4, M5 and M6. The boundary between...

AI Technical Summary

Problems solved by technology

This method only judges whether there is infarction in this area, but there are a large number of cases of ischemia without infarction in clinical practice. This method does not distinguish such cases, and many cases that need to be deducted will be missed, resulting in inaccurate scoring

And its infarction judgment method is the bilateral comparison of the regional mean value. When the infarct volume is smaller than the target area, the mean value changes slightly, and it is difficult to effectively distinguish it, resulting in a decrease in accuracy.

In addition, due to differences in normal CT values ​​of different structures of brain tissue, such as differences in gray matter and white matter CT values, it is difficult to truly distinguish infarction by simply comparing the mean difference

Images