Endoscopic Target Part Identification via Feature Similarity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing surgery assistance apparatuses struggle to track target part images during endoscopic surgery due to movement out of frame, changes in image-capturing parameters, concealment by body parts or water, and the non-rigid nature of internal body parts.
Innovation Solution
A surgery assistance apparatus that calculates feature amounts from endoscopic images, computes similarity degrees between different images, and identifies target part images if the similarity degree meets a predetermined value, thereby improving tracking accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional tracking techniques (such as optical flow) are used to track the target part image, then the tracking process can be performed, but the tracking accuracy deteriorates due to movement out of frame, changes in image-capturing parameters, concealment by body parts, and the non-rigid nature of internal body parts
Solution Approach 1:
The invention extracts multiple types of features (color, texture, shape) from the target part image and compares them across different images to identify the target part even when it moves or changes appearance. This multi-parameter feature extraction and comparison approach resolves the contradiction by maintaining identification accuracy despite changes in image-capturing parameters and target part position.
2Productivity
If the endoscope is inserted and extracted to perform surgery, then the surgical procedure can be completed, but the target part image cannot be tracked due to movement out of frame and changes in image-capturing distance and angle
Solution Approach 1:
The system continuously extracts features from each captured image and compares them with previously identified target part features, providing feedback to maintain tracking of the target part even as the endoscope moves. This feedback mechanism ensures that the target part identification is maintained throughout the surgical procedure despite changes in position and viewing angle.
3Ease of operation
If the target part is concealed by parts of the human body other than the target part (such as intestinal folds, intestinal mucus) or by water ejected from the endoscope, then the surgical view is obtained, but the target part image cannot be tracked
Solution Approach 1:
The system performs preliminary feature extraction and stores reference features of the target part before it becomes concealed. When the target part is temporarily hidden by body parts or water, the system uses the pre-extracted features to maintain identification, allowing tracking to resume immediately when the target becomes visible again without losing reliability.
Data Source
AI summary
Provided are a surgery assistance apparatus, a surgery assistance method, and a computer readable recording medium that includes a surgery assistance program recorded thereon that improve the accuracy of endoscopic surgery by identifying a target part image. A surgery assistance apparatus 1 includes a feature amount calculation unit 2 that calculates, from a human-body internal image captured using an endoscope, a feature amount of a target part image corresponding to a target part, a similarity degree calculation unit 3 that calculates a similarity degree of the features amount between different human-body internal images, and an identification unit 4 that identifies the target part image in each of the different human-body internal images if the similarity degree is greater than or equal to a predetermined value.