Abstract: A method for adding prediction results as training data using an AI prediction model, comprises the steps of: arranging an image having a first opposing attribute, the image being from an actual captured image, on a two-dimensional plane so as to be face-to-face with respect to a specific point on the two-dimensional plane; arranging an image having a second opposing attribute, on the two-dimensional plane so as to be orthogonal, with respect to the specific point, to the image having the first opposing attribute; receiving image data of the first and second attributes respectively arranged on the two-dimensional plane; by means of first and second AI prediction models, predicting respective results for the image data of the first and second attributes; converting the respective prediction results into X, Y coordinates, and cumulating and storing same; and selecting data to be included in training by analyzing the predicted results, and adding the selected data as training data.

Abstract: Proposed is a colonoscopy area indication method that includes loading a colonoscopy image analysis model and setting an analysis condition of the analysis model by a model loading/condition setting part, and initializing an analysis screen and displaying a picture of a normal colon by a controller, and preprocessing, by an image preprocessing part, a colonoscopy image received through an image receiving part so that subsequent image analysis is smoothly performed, and analyzing the preprocessed colonoscopy image by an image analysis part by using the image analysis model based on AI, and detecting and indicating, on the basis of a result of analysis by the image analysis part, at least one selected from a group of an examination area, a diverticulum area, and a lesion area in the colonoscopy image, and providing, by the controller, the result of analysis performed by the image analysis part.

Abstract: Proposed is a gastroscopy area indication method that includes loading a gastroscopy image analysis model and setting an analysis condition of the analysis model by a model loading/condition setting part, initializing an analysis screen and displaying a picture of a normal stomach by a controller, preprocessing, by an image preprocessing part, a gastroscopy image received through an image receiving part so that subsequent image analysis is smoothly performed, analyzing the preprocessed gastroscopy image by an image analysis part by using the image analysis model based on AI, detecting and indicating, on the basis of a result of analysis by the image analysis part, at least one selected from a group of an examination area, a hernia area, and a lesion area in the gastroscopy image, and providing, by the controller, the result of analysis performed by the image analysis part.

Abstract: A method of visualizing an analysis result of an endoscopic image according to a first embodiment includes loading, by a model loading/condition setting unit, a plurality of analysis models and setting analysis conditions, analyzing, by an image analyzer, the image using the plurality of analysis models when the situation requires image analysis, extracting, by an AM extractor/generator, AMs of the plurality of analysis models and generating a CAM when an image analysis result satisfies a set analysis condition, generating an ACAM by applying weights to CAMs extracted in previous and current image frames, extracting a peak activation value from the ACAM by an activation value extractor, and extracting, by a closed curve extractor, a closed curve of an AM and drawing the closed curve on a corresponding analysis image when the peak activation value is greater than or equal to a threshold.

Abstract: A method for providing a food allergy treatment service using oral immunotherapy, includes: transmitting patient information input into a patient terminal to a medical staff terminal; generating oral immunotherapy guide information on the basis of the patient information; transmitting the oral immunotherapy guide information to the patient terminal and displaying the same; when symptom information about symptoms after protein intake is input to the patient terminal, transmitting the symptom information to the medical staff terminal; analyzing the symptom information to determine an adverse reaction grade; and modifying the oral immunotherapy guide information, transmitting the same to the patient terminal, and displaying the oral immunotherapy guide information.

Abstract: Disclosed is an artificial intelligence (AI)-based gastroscopic image analysis method including a step of configuring training data and training the plurality of image classification models, a step of observing or photographing respective parts of a stomach from the oral cavity and the laryngopharynx to the second part of the duodenum using a gastroscopic probe, a step of automatically classifying and recognizing the anatomical location of the stomach with respect to an image captured during photographing using the plurality of image classification models, automatically storing or reporting the location of a lesion, and a step of segmenting an image of a specific part in the captured gastroscopic image for each region by a region segmentation model and outputting the same number of segmented maps as the number of target classes as the result of segmentation.

Abstract: A neural network model training method and an apparatus for complex characteristic classification and common localization are proposed. In the method, a neural network model includes: a convolution layer for performing a convolution operation on an input image by using a convolution filter; a pooling layer for performing pooling on an output of the convolution layer; and class-specific fully connected layers respectively corresponding to classes into which complex characteristics are classified and outputting values obtained by multiplying an output of the pooling layer by class-specific weights (Wfc(Tt). The method includes: (a) inputting the input image to the convolution layer; (b) calculating class-specific observation maps for respective classes on the basis of the output of the convolution layer; (c) calculating an observation loss (Lobs) common to the classes on the basis of the class-specific observation maps; and (d) back-propagating a loss based on the observation loss to the neural network model.

Abstract: A method of visualizing an analysis result of an endoscopic image according to a first embodiment includes loading, by a model loading/condition setting unit, a plurality of analysis models and setting analysis conditions, analyzing, by an image analyzer, the image using the plurality of analysis models when the situation requires image analysis, extracting, by an AM extractor/generator, AMs of the plurality of analysis models and generating a CAM when an image analysis result satisfies a set analysis condition, generating an ACAM by applying weights to CAMs extracted in previous and current image frames, extracting a peak activation value from the ACAM by an activation value extractor, and extracting, by a closed curve extractor, a closed curve of an AM and drawing the closed curve on a corresponding analysis image when the peak activation value is greater than or equal to a threshold.

Abstract: Disclosed is an artificial intelligence (AI)-based gastroscopic image analysis method including a step of configuring training data and training the plurality of image classification models, a step of observing or photographing respective parts of a stomach from the oral cavity and the laryngopharynx to the second part of the duodenum using a gastroscopic probe, a step of automatically classifying and recognizing the anatomical location of the stomach with respect to an image captured during photographing using the plurality of image classification models, automatically storing or reporting the location of a lesion, and a step of segmenting an image of a specific part in the captured gastroscopic image for each region by a region segmentation model and outputting the same number of segmented maps as the number of target classes as the result of segmentation.