Fundus Image Optic Cup and Disc Delineation with Deep Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Marking the optic cup and optic disc in fundus maps is challenging due to unclear boundaries and color variations, requiring significant time and expertise, especially in high myopia cases, hindering efficient risk assessment by ophthalmologists.
Innovation Solution
An image processing system utilizing a deep learning model trained on pre-collected fundus maps to recognize and mark optic cup and optic disc outlines, with an option for ophthalmologist modification, reducing manual marking time and improving interpretation efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual marking by ophthalmologists is used, then marking accuracy and quality are improved, but time consumption increases significantly
Solution Approach 1:
The system performs preliminary action by automatically detecting and marking optic cup and optic disc outlines before the ophthalmologist conducts manual marking. The deep learning model processes the fundus map and generates initial markings that serve as a foundation, reducing the time and effort required for the subsequent manual verification and adjustment process.
2Loss of time
If deep learning model automation is used, then time consumption is reduced, but marking accuracy may deteriorate
Solution Approach 1:
The system implements feedback by allowing the ophthalmologist to review, modify, and correct the automatically generated markings. The interface enables real-time adjustment of the outlined regions, and the system incorporates these corrections to improve accuracy. This feedback loop ensures that the final markings meet clinical standards while maintaining the time-saving benefits of automation.
3Measurement precision
If manual marking is performed for high myopia cases, then marking accuracy is maintained, but time consumption increases significantly
Solution Approach 1:
For high myopia cases, the system performs preliminary action by using the deep learning model to generate initial markings before manual verification. The model is trained to handle various fundus conditions including high myopia, providing a head start on the marking process that significantly reduces the time ophthalmologists need to spend on these challenging cases while maintaining accuracy through professional review.
4Measurement precision
If auxiliary marking function is added, then interpretation quality is improved, but system complexity increases
Solution Approach 1:
The system uses an intermediary approach by introducing a deep learning model as a mediator between the fundus map and the marking process. This intermediary component automatically performs the complex task of optic cup and optic disc detection, while the ophthalmologist's role is reduced to verification and adjustment. This mediation simplifies the overall workflow without requiring complex manual tracing tools.
Data Source
AI summary
Provided is an image processing system and method for analyzing an optic cup and optic disc. The system and method include the following steps. A processor obtains a fundus image. The processor recognizes the optic disc and optic cup in the fundus image using an image recognition model and marks the outlines of the optic disc and optic cup. An interpretation mode is provided so that interpretation data may be generated for risk assessment according to the outlines of the optic disc and optic cup as marked. The image recognition model is a deep learning model trained by a large amount of pre-collected fundus maps. The outlines of the optic cups and optic discs in the fundus images have been marked in advance.