Real-Time Endoscopic Navigation With AI Segmentation and 3D Model Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional endoscopic technology and imaging approaches fail to provide thorough, real-time visualization in anatomically constrained regions like the spinal canal, leading to increased risks of accidental damage to sensitive structures due to insufficient depth perception and lack of quantitative metrics for bone removal.
Innovation Solution
A system integrating endoscopic imaging with machine learning-based segmentation and patient-specific 3D model matching, utilizing a dual-camera stereoscopic vision module for enhanced depth perception and spatial orientation, and a neural network for real-time anatomical structure identification, enabling precise surgical navigation and guidance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Length of moving object
If conventional endoscopic technology is used, then minimally invasive access is achieved, but real-time visualization in anatomically constrained regions is insufficient
Solution Approach 1:
The patent transforms 2D endoscopic images into 3D spatial understanding by matching images with a 3D anatomical model. The system projects 3D model surfaces onto 2D image planes and compares them with actual endoscopic images, enabling depth perception and spatial orientation without changing the physical size of the endoscope.
Solution Approach 2:
The patent introduces a 3D anatomical model as an intermediary between the endoscopic camera and the surgeon's understanding. The model serves as a reference framework that bridges the gap between limited 2D visual information and the need for comprehensive 3D spatial awareness in constrained anatomical regions.
2Illumination intensity
If magnification tools such as operating microscopes or head-mounted loupes are used, then visibility is enhanced in straightforward cases, but detailed views around corners or through narrow canals cannot be obtained
Solution Approach 1:
The patent creates a virtual copy of the 3D anatomical structure through computer-generated modeling from preoperative scans. This virtual model can be viewed from any angle and provides comprehensive anatomical information that complements the limited direct visual field of the endoscope, enabling surgeons to understand structures around corners without physically repositioning the instrument.
3Productivity
If surgeons perform tissue removal blindly, then surgical procedure can continue, but the likelihood of accidental damage to nerves or other sensitive structures increases
Solution Approach 1:
The patent implements real-time feedback by continuously comparing endoscopic images with the 3D anatomical model and providing navigational guidance. The system offers quantitative metrics such as distance to critical structures and volumetric measurements, enabling surgeons to make informed decisions about tissue removal while maintaining procedural continuity.
4Loss of information
If conventional imaging modalities are used, then basic visualization is provided, but quantitative metrics for bone removal and correlation with preoperative findings are not available
Solution Approach 1:
The patent creates a multi-functional system that combines preoperative planning, real-time navigation, and quantitative measurement capabilities within a single 3D model framework. The same 3D model used for surgical planning also serves as a reference for real-time instrument localization and provides automated measurements of anatomical structures and removed tissue volume.
Data Source
AI summary
A method for real-time surgical navigation, including: capturing an intraoperative image stream from a distal portion of an endoscopic instrument inserted into a patient's body; processing the intraoperative image stream with a machine learning-based segmentation algorithm configured to identify anatomical structures; matching segmented images to a patient-specific three-dimensional (3D) model of the anatomy; and outputting navigational data indicating the position of the endoscopic instrument relative to the anatomical structures.