MR Machine learning based classification of primary brain tumor and brain metastasis using multimodal MR images

摘要

The present invention relates to a method for classifying primary brain cancer brain metastasis using a deep learning network-based multi-MR image, which configures a deep learning network and acquires an MR image for learning, and then images the MR image for learning on the deep learning network repeatedly learning the correlation between and image features; when the MR image of the patient is obtained, extracting image features of the MR image of the patient through the deep learning network and calculating a detailed brain cancer volume to calculate a feature vector; and determining and notifying the brain cancer type corresponding to the image characteristic by using a classifier in which the correlation between the characteristic vector for the image characteristic and the brain cancer type is predefined.