Fundus Image Processing for Non-Perfusion Area Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies face challenges in accurately predicting non-perfusion areas in fundus images, which are regions of the retina with little to no blood flow, often due to capillary vascular bed obstruction, limiting effective diagnosis and treatment of conditions like diabetic retinopathy and retinal vein occlusion.
Innovation Solution
An image processing method that involves performing first and second image processing steps to extract non-perfusion area candidates from fundus images, followed by predicting and displaying these areas using an image processing device and display device, enhancing vascular structures for precise identification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If single image processing method is used to extract non-perfusion area candidates, then processing time is reduced, but prediction accuracy deteriorates
Solution Approach 1:
The patent combines multiple image processing methods (first and second image processing) to extract non-perfusion area candidates from fundus images. By merging the results of different processing approaches, the system achieves more accurate prediction of non-perfusion areas while maintaining reasonable processing time through automated parallel execution of multiple algorithms.
2Measurement precision
If multiple image processing steps are performed to extract non-perfusion area candidates, then prediction accuracy is improved, but device complexity increases
Solution Approach 1:
The patent segments the image processing task into distinct first and second image processing steps, each extracting non-perfusion area candidates using different methodologies. This segmentation allows for systematic comparison and combination of results, improving overall prediction accuracy while organizing complexity into manageable, modular processing stages.
Solution Approach 2:
The patent introduces an intermediary step that combines the results from first and second image processing methods. This intermediary combination process acts as a mediator that integrates multiple processing outcomes to produce the final predicted non-perfusion areas, thereby improving accuracy while managing processing complexity through structured intermediate results.
3Productivity
If automated image processing is used to predict non-perfusion areas, then diagnostic efficiency is improved, but measurement precision may deteriorate due to algorithm limitations
Solution Approach 1:
The patent implements a feedback mechanism where the results from first and second image processing methods are compared and combined. This feedback loop allows the system to refine its predictions by validating results across multiple processing approaches, thereby maintaining high diagnostic efficiency while improving prediction accuracy through automated cross-validation.
Data Source
AI summary
An enhancement image processing section performs enhancement image processing on a fundus image of a subject eye to enhance vascular portions (304). A prediction processing section predicts a non perfusion area in the fundus image that has been subjected to the enhancement image processing (306 to 312). A generation section generates a non perfusion area candidate image (314).