AIArtificial Intelligence based gastroscope image analysis method

摘要

The invention relates to a gastroscope image analysis method based on artificial intelligence (AI). According to the invention, a gastroscope image analysis method based on artificial intelligence (AI),The stage and method of constructing multiple image classification models of gastroscopic photography based on artificial intelligence;Organize the respective learning data of multiple image classification models, and learn the stages and of multiple image classification models;The stage of observing or photographing each part of the gastrointestinal tract from the mouth and throat to the second part of the duodenum or from the second part of the duodenum to the mouth and throat by using the gastroscope probe;For the images taken during shooting, the above multiple image classification models are used to automatically classify and identify the gastric position,Automatically save or inform the location of the lesion,Verify whether the images of the recommended parts are taken and saved during gastroscopy;In the gastroscopic images, the images of specific parts are segmented according to the regional segmentation model as the segmentation results, including the stage of outputting the same number of segmentation maps as the number of target classes.