Endoscopic Navigation Training With Neural Network Guidance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Endoscopic procedures in human cavities are challenging due to reliance on medical personnel experience, inconsistent training methods, and the difficulty in accurately navigating endoscopes through complex and similar-looking structures, leading to potential examination errors and inefficiencies.
Innovation Solution
An image processing device using a supervised single-pass neural network model assists endoscope navigation by detecting anatomic reference positions in real-time, providing visual guidance without external devices, and incorporating proximity suppression logic to enhance prediction confidence.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If medical personnel rely on experience to navigate the endoscope through the human cavity, then the examination can be performed, but the navigation accuracy is insufficient and examination errors occur
Solution Approach 1:
The patent replaces the mechanical navigation method (relying on medical personnel's experience and manual control) with an automated image processing system. The system captures images from the endoscope, processes them to identify anatomical structures and the endoscope's position, and automatically determines navigation status, thereby eliminating human error in navigation accuracy assessment
Solution Approach 2:
The patent introduces an intermediary system (image processing device with neural network) between the endoscope and the medical personnel. This intermediary automatically analyzes images, identifies anatomical references, and determines the endoscope's position, serving as a mediator that enhances navigation accuracy without requiring direct human intervention in the critical navigation assessment
2Ease of manufacture
If training is based on a phantom model, then training can be conducted, but the training does not reflect actual clinical conditions and tissue movement
Solution Approach 1:
The patent creates a digital copy (virtual model) of the anatomical structure from the physical phantom model. The image processing system analyzes images captured during phantom model training and generates a corresponding virtual representation, allowing the training scenario to be replicated and analyzed computationally while maintaining the simplicity of the physical model
Solution Approach 2:
The patent transitions the training evaluation from a single physical dimension (phantom model) to multiple dimensions by adding a digital/virtual dimension. The system processes images from the phantom model and creates a computational representation that can be analyzed, stored, and reviewed in different dimensions, enhancing the training's adaptability and realism while maintaining the ease of using physical phantoms
3Reliability
If the endoscope is retracted and re-inserted to examine different parts, then comprehensive examination can be achieved, but the examination time increases
Solution Approach 1:
The patent implements a feedback mechanism where the image processing system continuously monitors the endoscope's position and the examined areas in real-time. The system provides immediate feedback to the medical personnel about the current location and examination status, enabling more efficient navigation and reducing unnecessary retraction and re-insertion operations while maintaining examination completeness
Solution Approach 2:
The patent performs preliminary analysis of the anatomical structure and planned examination path using image processing before the actual examination begins. The system pre-identifies key anatomical references and potential examination areas, allowing the medical personnel to plan the examination route more efficiently and reduce the time required for retraction and re-insertion operations
Data Source
AI summary
A training system for an endoscopic procedure, the training system including an endoscope having an image sensor configured to generate images; an image processing device including a housing and a processing circuit in the housing, the processing circuit including a processor and memory, the memory having graphical user interface (GUI) logic and a trained single-pass neural network model, the processor being configured to process the neural network model and the GUI logic to present a GUI; and a physical model including cavities corresponding to interconnected human cavities. The GUI is configured to receive a user input corresponding to a training mode or an exam mode. In the training mode the GUI is configured to present navigation feedback, and in the exam mode the GUI is configured to not present navigation feedback.