AI Endoscopic Lesion Detection With Blind Spot Notification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Endoscopic procedures for diagnosing stomach cancer are labor-intensive, time-consuming, and prone to variability due to factors like experience and fatigue, with challenges in analyzing vast amounts of images and potential blind spots, leading to inefficiencies and inaccuracies in lesion detection.
Innovation Solution
An AI-assisted endoscopy system that uses pre-learned endoscopic image data and medical information to automatically detect lesions in real time, applying deep learning models to enhance malignancy degree diagnosis accuracy and efficiency, and includes modules for lesion detection, classification, and notification of search completion and blind spots.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional endoscopic procedures are performed manually by specialists, then diagnostic expertise and flexibility are maintained, but the process becomes labor-intensive, time-consuming, and prone to inter- and intra-observer variability
Solution Approach 1:
An AI-based image processing system serves as an intermediary between the endoscopic procedure and diagnosis. The system automatically analyzes endoscopic images to detect lesions, reducing manual labor and time consumption while maintaining diagnostic accuracy through advanced image recognition algorithms
Solution Approach 2:
The manual mechanical process of specialist review is replaced with an automated AI-based image processing system. The AI algorithm processes endoscopic images to detect and classify lesions, eliminating inter- and intra-observer variability while improving diagnosis efficiency
2Reliability
If specialists review hundreds to thousands of endoscopic images manually, then comprehensive lesion detection is attempted, but the vast amount of images makes analysis and confirmation difficult and time-consuming
Solution Approach 1:
The AI-based image processing system performs self-service by automatically analyzing endoscopic images to detect lesions without requiring manual review of hundreds or thousands of images. The system processes images in real-time during the endoscopic procedure, significantly reducing analysis time while maintaining reliable lesion detection
Solution Approach 2:
The AI system performs preliminary analysis of endoscopic images during the procedure itself, identifying and flagging potential lesions before final diagnosis. This preliminary detection allows specialists to focus only on suspicious areas rather than reviewing all images manually
3Device complexity
If conventional endoscopy is used without AI assistance, then simple procedures are sufficient, but blind spots and inability to observe certain lesion types (e.g., Borrmann type 4, ulcerative lesions) occur
Solution Approach 1:
The AI-based image processing system provides multi-functional capability by detecting various types of lesions including Borrmann type 4 and ulcerative lesions that are difficult to observe with conventional endoscopy. The system analyzes images from multiple angles and patterns, eliminating blind spots while maintaining ease of use
Data Source
AI summary
An AI-based endoscopic diagnostic aid system includes: an endoscope module providing an endoscopic image of internal organs of the body of a patient; an input module configured to be capable of inputting arbitrary medical information about the patient; a control module which analyzes the endoscopic image provided from the endoscope module through a pre-stored image processing program to detect lesion information, matches the detected lesion information with the medical information input from the input module through a pre-stored lesion diagnosis program while generating at least one diagnosis information of malignancy and malignancy probability corresponding to the matching result, and outputs a preset notification signal according to the lesion information and the diagnosis information; and a notification module which visually displays on an arbitrary screen according to the notification signal output from the control module.