Breast cancer pathology image mitosis nucleus automatic segmentation method

Abstract

The invention discloses a breast cancer pathology image mitosis nucleus automatic segmentation method. A foreground pixel and a background pixel can be roughly estimated according to a maximum response of LoG filters of different scales of a BR graph. The color distribution of the foreground pixel and the background pixel can be described by Gaussian Mixed Models GMM, which are trained respectively according to the current foreground pixel and the current background pixel. An s-t graph G=(V,E) can be established, and the new estimation of the foreground and the background can be realized by calculating the maximum flow of the graph G, and then the GMM training and the Graph-Cut segmentation can be repeated, till the convergence is realized, or the predetermined iterations can be satisfied. The morphology open operation of the segmentation result can be carried out, and the final segmentation can be completed. The breast cancer pathology image mitosis nucleus automatic segmentation method is advantageous in that during the segmentation, the probability of the single pixel color characteristic belonging to the foreground or the background can be considered, and the relation between the adjacent pixels can be considered, and therefore the smoothness of the segmentation results can be guaranteed, and the precision can be improved.

Description

technical field [0001] The invention belongs to the technical field of image processing, in particular to a method for automatically segmenting mitotic nuclei in pathological images of breast cancer. Background technique [0002] According to statistics from the Ministry of Health, breast cancer accounts for a large proportion of cancers in women in my country, and has become one of the biggest killers that endanger women's lives. The age of onset ranges from 20 to 70 or 80 years old, and the incidence It tends to increase with age. At present, most breast cancer patients in our country are concentrated in the age group of 45-55, and the incidence rate accounts for 7-10% of all kinds of malignant tumors in the whole body, which is second only to the incidence rate of uterine cancer in women. At present, the etiology of breast cancer is not completely clear, and its early detection and early diagnosis are very important. [0003] Mitotic count is an important indicator in th...

AI Technical Summary

Problems solved by technology

A large number of image segmentation methods such as threshold segmentation, watershed segmentation, morphological methods, and LoG-based blob detection have been used to segment cell nuclei. However, these algorithms are too simple, and often segment a large number of mitotic nuclei (positive samples) while segmenting The non-mitotic part (negative samples) of the average positive and negative samples reaches 1:100, and the unbalanced distribution of positive and negative samples brings great difficulties to the subsequent classifier training.

Images