Intelligent detection method of mitotic figures in gastrointestinal stromal tumors based on deep learning

Abstract

A method for intelligent detection of mitosis in gastrointestinal stromal tumors based on deep learning: preprocessing the acquired pathological images stained with hematoxylin-eosin; using EfficientDet-D0 as a deep learning detection model and performing training; using U-Net is used as a deep learning segmentation model and trains the deep learning segmentation model; constructs a deep learning classification model; trains a deep learning classification model; uses the trained deep learning detection model to stain the subjects' hematoxylin-eosin Pathological images are detected; the pathological images are segmented using the deep learning segmentation model, and the segmented results are detected; the mitotic detection results based on the deep learning detection model are compared with the mitotic detection results based on the deep learning segmentation model, Get the final classification result. The invention realizes the judgment of the risk degree of gastrointestinal stromal tumor by analyzing the input hematoxylin-eosin stained image and detecting the number of mitotic figures therein.

Description

technical field [0001] The invention relates to a method for detecting mitoses in gastrointestinal stromal tumors. In particular, it relates to a deep learning-based intelligent detection method for mitotic figures in gastrointestinal stromal tumors. Background technique [0002] Gastrointestinal stromal tumor (GIST) is a type of tumor originating from the mesenchymal tissue of the gastrointestinal tract, accounting for the majority of gastrointestinal mesenchymal tumors, and is the most common abdominal soft tissue malignant tumor. Gastrointestinal stromal tumors are most commonly found in the stomach (50%-60%), followed by the small intestine (30%-35%), colon and rectum (5%), esophagus (<1%), and, to a lesser extent, outside the gastrointestinal tract ( mesentery, omentum and retroperitoneum; <5%). The mean age at diagnosis was 63 years, with no gender differences. Patients with gastrointestinal stromal tumors often have no characteristic symptoms because the tumo...

AI Technical Summary

Problems solved by technology

However, due to the characteristics of single category and extremely unbalanced foreground-background of pathological pictures, a method for detecting and counting mitotic figures in hematoxylin-eosin stained pathological section images is needed that combines deep learning detection methods, segmentation methods and classification methods

Images