AI(Artificial Intelligence) based gastroscope image analysis method

摘要

The present invention relates to an artificial intelligence (AI)-based gastroscopic image analysis method. The method for analyzing a gastroscopy image based on artificial intelligence (AI) according to the present invention includes the steps of constructing a plurality of image classification models based on artificial intelligence for a gastroscopy image; Configuring each of the training data for a plurality of image classification models to train a plurality of image classification models; Observing or photographing each part of the stomach from the oral cavity and the throat to the second part of the duodenum or from the second part of the duodenum to the oral cavity and the throat using a gastroscopy probe; With respect to the image taken during shooting, the anatomical position of the stomach is automatically classified and recognized using the plurality of image classification models, and the location of the lesion is automatically saved or notified, and shooting and saving are recommended in gastroscopy. Verifying whether or not an image of the portion to be taken has been captured; And dividing an image of a specific portion of the captured gastroscopy images for each region by a region division model, and outputting the same number of division maps as the number of target classes as a division result.