Enterophthisis auxiliary diagnosis system and method based on deep learning

Abstract

The invention discloses an enterophthisis auxiliary diagnosis system and method based on deep learning. The system comprises a colonoscopy image automatic collection subsystem, a client and a database. The colonoscopy image automatic collection subsystem is used for collecting colonoscopy images, and the client includes an image preprocessing module, a convolutional neural network module, an imagedisplay module, and a colonoscopy report output module. The convolutional neural network module comprises an image eligibility discriminant submodel, an image part discriminant submodel, and a discrimination submodule for determining whether an image contains a focus of infection. The database is used for saving a sample set for training the convolutional neural network, and storing the information outputted by the colonoscopy report output module. The system utilizes image recognition technology to monitor an endoscope video in real time, automatically collects images containing key organ parts and suspicious lesion areas, and saves them in the database after global selection according to a weighting algorithm. The system can extract the most valuable images from the global video and provide more reliable and efficient support for doctor diagnosis.

Description

technical field [0001] The invention belongs to the technical field of image recognition, and relates to a medical endoscope image recognition system and method, in particular to a deep learning-based auxiliary diagnosis system and method for intestinal tuberculosis. Background technique [0002] With the continuous development and maturity of deep learning algorithms, it has been gradually used in the field of medical image analysis. Endoscopic images are an important basis for doctors to analyze patients' digestive tract diseases. In recent years, a variety of methods for screening and diagnosing lesions using deep convolutional neural network models have been developed, which is of great clinical significance in the current colonoscopy diagnosis system. [0003] Intestinal tuberculosis mainly involves the ileocecal region and its adjacent colon, and its colonoscopic appearance does not show segmental distribution, and the ulcers are mostly horizontal, superficial and irre...

AI Technical Summary

Problems solved by technology

[0005] The present invention mainly solves the problem that the traditional colonoscopy report system relies on manual image collection, which is prone to missing blind spots and suspicious lesion areas in site inspection. It uses image recognition technology to monitor endoscopic video in real time, and automatically collects key organ sites and suspicious lesion areas. After the images are selected globally according to the weighted algorithm, they are saved in the colonoscopy report system

Images