|
1. A method of predicting a region-specific cerebral cortical contraction rate on basis of a computed tomography (CT) image, the method comprising: a deep learning step of training a deep learning network, by selecting and using CT images of a plurality of patients and segmentation information, about a correlation between the CT images and the segmentation information; a feature extraction step of extracting, on basis of each segmentation information, semantic feature information corresponding to each of the CT images; a machine learning step of additionally obtaining a region-specific cerebral cortical contraction rate corresponding to each semantic feature information and then training a machine learning model about a correlation between the semantic feature information and the region-specific cerebral cortical contract rate; a segmentation step of, when an image to be analyzed is input, obtaining segmentation information corresponding to the image to be analyzed through the deep learning network; and a prediction step of extracting semantic feature information corresponding to the image to be analyzed on basis of the segmentation information and then predicting and notifying a region-specific cerebral cortical contraction rate corresponding to the semantic feature information through the machine learning model.
|
