Deep learning-based intelligent detection method for nuclear division images in gastrointestinal stromal tumor

Abstract

The invention discloses a deep learning-based intelligent detection method for nuclear division images in gastrointestinal stromal tumor. The method comprises the following steps: preprocessing an obtained hematoxylin-eosin staining pathological image; taking EfficientDet-D0 as a deep learning detection model, and carrying out training; using U-Net as a deep learning segmentation model, and training the deep learning segmentation model; constructing a deep learning classification model; training the deep learning classification model; detecting the hematoxylin-eosin staining pathological imageof the testee by using the trained deep learning detection model; segmenting the pathological image by using a deep learning segmentation model, and detecting the segmented result; and comparing thenuclear division images detection result based on the deep learning detection model with the nuclear division images detection result based on the deep learning segmentation model to obtain a final classification result. According to the invention, the input hematoxylin-eosin staining image is analyzed, and the number of nuclear division images is detected, so that the judgment on the risk degreeof gastrointestinal stromal tumor is realized.

Description

technical field [0001] The invention relates to a method for detecting mitotic figures in gastrointestinal stromal tumors. In particular, it relates to an intelligent detection method for mitotic figures in gastrointestinal stromal tumors based on deep learning. Background technique [0002] Gastrointestinal stromal tumor is a type of tumor originating from the mesenchymal tissue of the gastrointestinal tract, which accounts 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 a small part outside the digestive tract ( Mesentery, omentum, and retroperitoneum; <5%). The mean age at diagnosis was 63 years, with no gender difference. Patients with gastrointestinal stromal tumors often have no characteristic symptoms because the tumor is lo...

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