Digestive tract tumor infiltration layer identification method and device, computer equipment and medium

Abstract

The embodiment of the invention discloses a digestive tract tumor infiltration layer identification method and device, computer equipment and a medium. The method comprises the following steps: inputting a received ultrasonic endoscope image into a pre-trained first neural network model to obtain an effective area of the ultrasonic endoscope image; inputting the effective area into a pre-trained second neural network model to respectively obtain a mucosa layer area, a mucosa muscle layer area, a submucosa layer area, an innate muscle layer area and a serosa layer area of the digestive tract wall in the ultrasonic endoscope image; sequentially generating minimum circumscribed rectangles of the mucosa layer area, the mucosa muscle layer area, the submucosa layer area, the innate muscle layer area and the serosa layer area; and identifying the alimentary canal tumor infiltration layer area in the ultrasonic endoscope image according to each minimum circumscribed rectangle. According to the method, the digestive tract tumor infiltration layer area of the ultrasonic endoscope image is accurately identified by adopting an artificial intelligence technology, and the accuracy of identifying the digestive tract tumor infiltration layer is improved.

Description

technical field [0001] The invention relates to the field of medical assistance technology, in particular to a method, device, computer equipment and medium for identifying an infiltration layer of alimentary canal tumors. Background technique [0002] The wall infiltration layers of gastrointestinal tumors can be divided into superficial layer and deep layer. The superficial layer includes: mucosal layer, muscularis mucosae and submucosa layer, and the deep layer includes: muscularis propria and serosa layer. Endoscopic ultrasound (EUS) is a digestive tract inspection technology that combines endoscopy and ultrasound. A miniature high-frequency ultrasound probe is placed on the top of the endoscope, and the endoscope is inserted into the body cavity to directly observe the digestive tract mucosa At the same time as the lesion, the ultrasound under the endoscope is used to scan in real time to obtain the hierarchical structure of the digestive tract wall, so as to identify t...

AI Technical Summary

Problems solved by technology

[0004] Embodiments of the present invention provide a method, device, computer equipment and medium for identifying infiltrating layers of digestive tract tumors, which are used to solve the technical problem of low recognition rate of infiltrating layers of digestive tract tumors in ultrasonic endoscopic images in the prior art

Images