Pathological image prognosis feature weight calculation method and system

Abstract

The invention discloses a pathological image prognostic feature weight calculation method and system, and the method comprises the steps of obtaining the features of a to-be-processed digital pathological image through the analysis of the to-be-processed digital pathological image, enabling a pathological report and clinical information corresponding to pathological image-text data with high similarity, and converting the pathological image-text data into text data; establishing a prognostic feature extraction model based on a graph convolution network, and obtaining prognostic features and attribute values corresponding to the prognostic features from the text data converted from the pathological image-text data by utilizing the prognostic feature extraction model; and finally, performingsurvival analysis or risk index calculation on the post-features and the attribute values corresponding to the prognosis features to obtain prognosis feature weight values of the pathological images,thereby endowing weights to various pathologies and clinical features of the pathological images similar to the digital pathological images to be processed for each digital pathological image to be processed, and improving the pathological analysis efficiency.

Description

technical field [0001] The invention belongs to the field of digital pathological image processing, and relates to a pathological image prognosis feature weight calculation method and system. Background technique [0002] With the wide popularization and application of digital pathological slice images, the automatic calculation of histopathological image analysis system has been gradually promoted and used in the analysis of pathological images. Histopathological images provide important information for staging and prognosis, and pathologists in It has been widely used in clinical practice. In the medical field, pathological analysis accurately determines pathological features is the "gold standard" of the medical analysis process, and is the cornerstone for doctors to correctly analyze patients. There are many different pathological features in the pathology report, such as cancer type, cancer subtype, anatomical site , tumor size, grade, TNM staging. These features play...

AI Technical Summary

Problems solved by technology

At present, some pathological image analysis methods often lack clear pathological test results as the basis, and are only based on speculation with insufficient evidence, mainly relying on the personal experience of pathologists, which cannot be combined with the existing experience of pathological image results. It takes a long time to train an excellent pathologist, and an experienced pathologist needs the accumulation of many different types of pathological analysis, which is difficult, and the results of many pathological analyzes currently available , relying on manual analysis, it is impossible to quickly and accurately obtain accurate pathological feature information from pathological images, and the accuracy of calculation and analysis results is low

Images