RNFL Defect Detection via Polar Coordinate Conversion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current glaucoma diagnosis methods are time-consuming and complex, making it difficult to rapidly detect defective zones in the retinal nerve fiber layer (RNFL) from fundus images.
Innovation Solution
A method involving fundus image acquisition, detection of the optic disc center, polar coordinate conversion, and image processing stages that include green channel extraction, brightness correction, blood vessel removal, noise reduction, and Hough transformation to detect atrophy zones in the RNFL.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple glaucoma diagnosis tests (tonometry, optic nerve test, visual field test, gonioscopy) are performed, then diagnostic accuracy is improved, but examination time and patient fatigue increase
Solution Approach 1:
The invention extracts and focuses on a specific diagnostic indicator (RNFL defect) from the complex set of glaucoma diagnosis tests. By using image processing to directly analyze and detect RNFL defects from fundus images, the system isolates one effective diagnostic method that can provide reliable glaucoma detection without requiring all the traditional multiple tests, thereby reducing examination time while maintaining diagnostic accuracy.
2Reliability
If traditional multiple glaucoma diagnosis procedures are used, then reliable diagnosis is achieved, but patient fatigue increases
Solution Approach 1:
The invention extracts the essential diagnostic information (RNFL defect detection) from the complex multi-test procedure and implements it through automated image processing. This allows reliable diagnosis to be achieved through a simpler, less invasive method that does not require patients to undergo multiple lengthy procedures, thereby improving patient comfort while maintaining diagnostic reliability.
3Productivity
If rapid RNFL defect detection is implemented through image processing, then examination time is reduced, but detection precision may be compromised
Solution Approach 1:
The invention replaces manual, time-consuming examination procedures with automated image processing systems. By using computer-based algorithms to analyze fundus images and detect RNFL defects, the system achieves rapid detection speed while maintaining or even improving precision through consistent, objective automated measurement rather than subjective manual assessment.
Solution Approach 2:
The invention transforms the diagnostic approach by changing from multiple physical tests to digital image analysis. By converting anatomical structures into digital images and applying image processing algorithms, the system enables rapid automated detection while maintaining precision through quantitative image analysis parameters such as RNFL thickness measurement and defect pattern recognition.
Data Source
AI summary
The present invention relates to a method for detecting a defective zone of a retinal nerve fiber layer (hereinafter, referred to as a RNFL), wherein a green channel image is extracted from an acquired fundus image, and polar coordinate conversion is conducted with reference to the center of a detected optic disc with regard to the image, thereby detecting a defective area of the RNFL.