Diaphragm Edge Detection in X-Ray Images Using Structural Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing computer-assisted detection methods for diaphragm edges in x-ray images are unreliable under adverse conditions, particularly when large implants are present or when diaphragm contrast is similar to the patient contrast, leading to incorrect or missed edge recognition.
Innovation Solution
The method involves providing the computer with information about the inner structure of the diaphragm, such as regular geometric shapes, and using derivative values and weighting factors to detect diaphragm edges by identifying edges that form specific angles and maintaining a minimum separation, thereby enhancing the accuracy of edge detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If computer-assisted detection methods are used to automatically detect diaphragm edges, then productivity is improved, but reliability deteriorates under disadvantageous conditions such as large implants or low contrast
Solution Approach 1:
The detection process is segmented into multiple stages: first detecting all potential image edges, then filtering these edges based on structural criteria (forming rectangles with specific angles and parallel sides), and finally identifying diaphragm edges from the filtered set. This multi-stage segmentation allows automatic detection to maintain high productivity while improving reliability through progressive filtering.
Solution Approach 2:
The patent introduces an intermediary filtering mechanism that uses known diaphragm structural properties (rectangular shape, 90-degree angles, parallel opposite sides) as intermediate criteria to bridge the gap between automatic edge detection and reliable diaphragm edge identification. This intermediary layer eliminates false detections from implants or other structures while preserving true diaphragm edges.
2Reliability
If manual input methods are used to enter diaphragm edge positions, then reliability is improved, but productivity deteriorates due to slow and laborious operation
Solution Approach 1:
The system performs self-service by automatically detecting and identifying diaphragm edges using computational algorithms that analyze image data and apply structural filtering criteria. This eliminates the need for manual user input while maintaining high detection accuracy, thereby resolving the contradiction between reliability and productivity.
3Device complexity
If simple edge detection algorithms are used, then device complexity is reduced, but measurement precision deteriorates when diaphragm contrast differs little from patient contrast
Solution Approach 1:
The patent applies preliminary action by first detecting all potential edges in the image, then using preliminary filtering based on known diaphragm structural properties (rectangular geometry, angle relationships, parallel sides) to identify which edges belong to the diaphragm. This preliminary structuring of the detection process improves measurement precision without requiring overly complex algorithms.
Solution Approach 2:
The system changes detection parameters dynamically by adjusting the stringency of structural criteria application based on image characteristics. When contrast is low, the algorithm relies more heavily on geometric structural parameters (angles, parallelism, rectangle formation) rather than purely on intensity-based edge detection, thereby maintaining precision without excessive complexity.
Data Source
AI summary
In a computer-assisted method for detecting diaphragm edges caused in an image by a diaphragm, the image and information about an inner structure of the diaphragm are provided to a computer. The computer first detects all image edges that are present in the image. Using the information about the inner structure of the diaphragm, it then determines the diaphragm edges from among the detected image edges.