Endoscopic Image Region Association with 3D Medical Coordinates
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing virtual endoscope systems do not effectively associate regions of interest from endoscopic images with three-dimensional X-ray CT images during diagnosis, leading to errors in image measurement and diagnosis support.
Innovation Solution
A program that acquires endoscopic images, extracts region-of-interest information, and associates it with three-dimensional medical images using position information in a coordinate system, allowing for efficient storage and retrieval of this information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a virtual endoscope simply displays a cross-sectional X-ray CT reconstructed image, then the three-dimensional structure can be visualized, but the association with region of interest from endoscopic images is not achieved
Solution Approach 1:
The patent introduces a coordinate system conversion mechanism as an intermediary between the endoscopic image coordinate system and the X-ray CT image coordinate system. This mediator enables accurate position mapping by converting coordinates from one system to the other, allowing precise association of regions of interest without direct complex integration of the two imaging systems.
Solution Approach 2:
The patent divides the association process into distinct segments: (1) acquiring endoscopic images and extracting regions of interest, (2) acquiring X-ray CT images and establishing coordinate systems, (3) converting coordinates between systems, and (4) associating the data. This segmentation allows each step to be handled independently, reducing overall system complexity while maintaining measurement precision.
2Loss of information
If region of interest information is extracted from endoscopic images, then diagnostic information can be obtained, but the position information in three-dimensional medical image coordinate system cannot be derived
Solution Approach 1:
The patent uses coordinate transformation as an intermediary process to bridge the gap between endoscopic image coordinates and X-ray CT coordinates. By introducing this transformation layer, the system can derive position information in the three-dimensional medical image coordinate system without requiring direct complex integration, thus maintaining information completeness while managing processing complexity.
Solution Approach 2:
The patent performs preliminary actions by pre-establishing coordinate systems for both endoscopic and X-ray CT images, and pre-defining the transformation relationships between them. This preliminary setup enables subsequent region of interest association to be performed more efficiently, reducing the complexity of real-time processing while ensuring no information is lost.
3Reliability
If endoscopic images and three-dimensional medical images are stored separately, then data management is simple, but association between the images cannot be achieved
Solution Approach 1:
The patent merges the endoscopic image data and X-ray CT image data by establishing associations through coordinate transformation and position information matching. This combining approach maintains the reliability needed for diagnosis support by ensuring the images are properly correlated, while managing data management complexity through structured association rules rather than complete data integration.
Data Source
AI summary
A program causes a computer to perform processing of: acquiring an endoscopic image obtained by capturing a subject using an endoscope; extracting region-of-interest information from the acquired endoscopic image; acquiring a three-dimensional medical image obtained by capturing an inside of a body of the subject using at least one of X-ray CT, X-ray cone beam CT, MRI-CT, and an ultrasonic diagnostic device; deriving position information in a coordinate system of the three-dimensional medical image specified by the region-of-interest information and the three-dimensional medical image; and storing the region-of-interest information and the three-dimensional medical image in association with each other based on the derived position information and a capturing time point of each of the endoscopic image and the three-dimensional medical image.