A hepatocellular carcinoma automatic grading method based on an SE-DenseNet deep learning framework and a multi-modal enhanced MR image

Abstract

The invention discloses a hepatocellular carcinoma automatic grading method based on an SE-DenseNet deep learning framework and an enhanced MR image. The method comprises the following steps of 1) collecting data; 2) preprocessing all hepatocellular carcinoma three-dimensional images with enhanced MR; 3) enhancing the training data; 4) based on the enhanced training data, training a hepatocellularcarcinoma grading prediction model, namely an SE-DenseNet network; and 5) carrying out hierarchical prediction on the test data by adopting the trained model, and evaluating the classification performance of the hepatocellular carcinoma hierarchical prediction model. According to the automatic pathological grading method for the hepatocellular carcinoma multi-modal enhanced MR image which is composed of the steps of image preprocessing, the image enhancement, the hepatocellular carcinoma multi-modal enhanced MR image classification, SE-DenseNet network training and SE-DenseNet network testing, the hepatocellular carcinoma automatic grading can be realized, and the problems of manpower consumption, time consumption and subjective difference existing in manual hepatocellular carcinoma grading can be solved.

Description

Technical field [0001] The present invention relates to the field of medical image processing, in particular to an automatic classification method of hepatocellular carcinoma based on SE-DenseNet deep learning framework and multimodal enhanced MR images. Background technique [0002] Globally, lung / liver / stomach and bowel tumors currently account for nearly half (46%) of all cancer deaths. Hepatocellular carcinoma (HCC), fibrous membrane carcinoma, cholangiocarcinoma, angiosarcoma and hepatoblastoma are called primary liver cancers. Among these primary liver cancers, hepatocellular carcinoma (HCC) is the third leading cause of cancer deaths worldwide Large tumors, more than 500,000 people are affected by this disease. In the grades of hepatocellular carcinoma, grades II and III account for the majority. The prognosis of liver cancer patients is related to the differentiation degree of liver cancer and the treatment plan of liver cancer. Therefore, the classification of liver ca...

AI Technical Summary

Problems solved by technology

[0003] The purpose of the present invention is to solve the problems of manpower, time consumption and subjectivity differences in the traditional manual grading of liver cells, and proposes a liver cell based on SE-DenseNet deep learning framework and multimodal enhanced MR images. automatic cancer grading method

Images