Virtual Fluorescent Fundus Image Generation via Color Feature Encoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Fluorescent fundus photography is time-consuming and uncomfortable due to the need for a mydriatic and fluorescent agent, posing risks such as nausea and anaphylactic shock, limiting its use despite high accuracy.
Innovation Solution
A method and device that transform a color fundus image into a fluorescent fundus image using a fundus image transformation model, eliminating the need for the fluorescent agent by encoding color features, thereby reducing wait time and discomfort.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If fluorescent fundus photography is performed using fluorescent agent, then diagnostic accuracy is improved, but test time and patient discomfort increase
Solution Approach 1:
The patent creates a virtual fluorescent fundus image by copying and transforming the color fundus image through encoding models, rather than capturing a real fluorescent image requiring chemical agents. This virtual copy replicates the diagnostic information of fluorescent imaging without the physical constraints, thereby maintaining diagnostic accuracy while eliminating waiting time and patient discomfort associated with actual fluorescent agent administration
2Measurement precision
If fluorescent fundus photography is performed using fluorescent agent, then diagnostic accuracy is improved, but patient comfort deteriorates
Solution Approach 1:
The patent generates a virtual fluorescent fundus image from the color fundus image through encoding transformation, eliminating the need for actual fluorescent agent injection. This virtual copying approach preserves all diagnostic information while completely avoiding the harmful effects of chemical agents including nausea, vomiting, allergies, and anaphylactic shock, thereby maintaining diagnostic accuracy while ensuring patient comfort
3Loss of time
If virtual fluorescent fundus image is generated through encoding, then test time is reduced, but image transformation complexity increases
Solution Approach 1:
The patent pre-trains the encoding model using paired color and fluorescent fundus images to learn the transformation mapping. This preliminary action prepares the model in advance, enabling rapid real-time transformation of color images to virtual fluorescent images during actual examinations. The complex transformation work is done beforehand during model training, not during patient examination, thereby reducing test time while managing transformation complexity through pre-computation
Data Source
AI summary
The disclosure provides a method for obtaining a fluorescent fundus image, the method including the following steps: receiving a to-be-detected color fundus image; obtaining a color feature of the to-be-detected color fundus image; encoding the color feature of the to-be-detected color fundus image according to a fundus image transformation model to transform the to-be-detected color fundus image into a to-be-detected fluorescent fundus image; and outputting the to-be-detected fluorescent fundus image.