Endoscope Navigation Guidance for Complete Digestive Tract Coverage
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing gastroscopy procedures are time-consuming and prone to incomplete examination of digestive tract sites, leading to potential missed lesions and increased workload for endoscopists, which can compromise patient safety and examination quality.
Innovation Solution
An auxiliary endoscopy system utilizing an anatomic-site identification module to map digestive tract sites, an attention degree recording module to sequence observations, and a guide module to optimize endoscope navigation, ensuring comprehensive and efficient examination.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If endoscopists perform gastroscopy manually without assistance, then they can detect suspicious areas, but it takes a long time and requires extensive experience leading to incomplete examination
Solution Approach 1:
The system enables self-service by having the endoscopy device automatically identify anatomical sites, determine observation sequences, and guide navigation without requiring endoscopist expertise. The AI system performs the examination guidance functions that would otherwise require years of training, allowing less experienced operators to achieve complete examinations.
Solution Approach 2:
The patent replaces the mechanical system of human expertise and manual navigation with an AI-based automated system. The neural network models substitute for the endoscopist's trained judgment, while automated guidance mechanisms replace manual skill-based navigation, enabling consistent and complete examinations regardless of operator experience level.
2Reliability
If endoscopists increase their workload to ensure complete coverage, then examination quality improves, but it leads to fatigue and reduced observation integrity
Solution Approach 1:
The system implements feedback mechanisms where the AI continuously monitors the endoscopy procedure, compares actual observations against the planned sequence, and provides real-time guidance. This feedback loop ensures complete coverage of all anatomical sites while maintaining efficient progression through the examination, preventing both omissions and unnecessary repetitions.
Solution Approach 2:
The system performs preliminary actions by pre-determining the optimal observation sequence and required examination steps before the procedure begins. The AI analyzes anatomical relationships and establishes a standardized examination pathway in advance, allowing the endoscopist to follow a predetermined efficient route rather than relying on real-time decision-making under pressure.
3Adaptability or versatility
If endoscopists use traditional manual examination methods, then they can adapt to individual patient conditions, but it results in repeated examinations and wasted resources
Solution Approach 1:
The system maintains adaptability through dynamic adjustment capabilities. The AI-generated examination sequence can be modified in real-time based on actual observations, patient anatomy variations, and clinical needs. The system adapts the standardized protocol to individual cases while ensuring complete coverage, preventing the need for repeated examinations due to missed sites.
4Measurement precision
If endoscopists rely on experience and memory, then they can identify suspicious areas, but it leads to missed examination sites and incomplete reports
Solution Approach 1:
The system creates a digital copy or model of the anatomical structure through AI-generated three-dimensional representations and virtual anatomical maps. This virtual model serves as a reference guide during the examination, providing precise identification of anatomical sites without requiring the endoscopist to rely on memory or experience. The digital twin approach simplifies the complexity by providing intuitive visual guidance.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The present application relates to an endoscope auxiliary examination system, a method, an apparatus, and a storage medium. The system comprises: an anatomical part identification module, used to perform frame-by-frame identification on collected endoscope examination images, determine every observation point part corresponding to each frame of the endoscope examination images, and map each observation point part to a preset site part; an attention recording module, used to determine a site part observation sequence according to a recorded observation time of each observation point part and an observation completeness of each site part determined by the anatomical part identification module; and a guidance module, used to determine a target site to be moved to, according to a current site part corresponding to a current frame endoscope examination image determined by the anatomical part identification module, the site part observation sequence and a preset physical depth of the current site part, and guiding endoscopic examination according to the current site part and the target site. According to the present application, an endoscopist is guided to operate an endoscope according to the most efficient route.