Automatic staging system based on mammary gland lymph node panoramic image calculation

Abstract

The invention discloses an automatic staging system based on mammary gland lymph node panoramic image calculation, and belongs to the technical field of medical image information processing. The method comprises: establishing two positive and negative samples required by the deep convolutional network according to the marked lymph node cancer metastasis region, training the deep convolutional network to obtain a model required by classification of the positive and negative samples, sending each block in the image tissue region into the model to obtain the probability of cancer metastasis at the position of the block, and establishing a probability calorific value graph; Secondly, extracting pathological omics characteristics according to a segmentation area of lymph node cancer metastasis,training a random forest classifier, constructing an automatic classification model of lymph node metastasis states, and constructing a staging system by integrating multiple lymph node cancer metastasis states of the same patient. The problems of time consumption, labor consumption, high error and the like when breast cancer staging is judged by pathology experts according to experience can be effectively solved.

Description

technical field [0001] The invention relates to an automatic staging system based on panoramic image calculation of breast lymph nodes, and belongs to the technical field of medical image information processing. Background technique [0002] Breast cancer is the malignant tumor with the highest incidence rate among women in the world, and its mortality rate is about 6%, second only to lung cancer with the highest mortality rate. According to the latest cancer statistics report of China and the United States, compared with other malignant tumors, breast cancer has a better prognosis. If early detection, early diagnosis, and early treatment can be achieved, the cure rate can reach more than 96%. Metastasis is the most important prognostic factor in cancer diagnosis and treatment. Lymph node cancer metastasis is the most common way of tumor metastasis. A phenomenon in which a tumor grows centered on a lymph node. The presence of cancer cells in the sentinel lymph nodes indica...

AI Technical Summary

Problems solved by technology

[0011] The technical problem to be solved in the present invention is to solve the problems of time-consuming, laborious, and large errors in traditionally relying on pathologists to judge breast cancer staging based on experience, and proposes an automatic staging system based on panoramic image calculation of breast lymph nodes, using deep learning methods to distinguish breast cancer In the pathological image of cancer panorama scanning, the normal area and the tumor area are different in pathological tissue structure, type complexity, and shape. There are many targets to be detected and segmented, and there are complex background, large noise, uneven staining, and imaging quality. It is not conducive to accurately judge the problem of breast cancer staging, and realize automatic and accurate staging of panoramic images of breast lymph nodes

Images