Method for analyzing retinal fundus images and related products

Abstract

The invention relates to a method for analyzing retinal fundus images and related products. Wherein the method includes: using a pre-trained arteriovenous segmentation model to process the retinal fundus image to obtain an arterial segmentation result and a vein segmentation result, wherein the arterial segmentation result includes a continuous arterial network, and the vein segmentation result includes A continuous vein network; acquiring position information of a positioning reference object in the retinal fundus image; and determining the arterial network and the position information according to the position information of the positioning reference object, the artery segmentation result, and the vein segmentation result The structural attribute information of the key points in the vein network. Through the technical solution of the present invention, using the continuous arterial network and venous network of the retinal fundus image, combined with the position information of the positioning reference object, the refined detection of the multi-dimensional structural attributes of the key points can be realized.

Description

technical field [0001] The present invention generally relates to the technical field of image processing. More particularly, the present invention relates to a method for analyzing retinal fundus images, as well as devices and computer-readable storage media for performing the aforementioned method. Background technique [0002] This section is intended to provide a background or context for the embodiments of the invention that are recited in the claims. The descriptions herein may include concepts that may be explored, but not necessarily concepts that have been previously thought of or explored. Therefore, unless otherwise indicated herein, what is described in this section is not prior art to the description and claims in this application and is not admitted to be prior art by inclusion in this section. [0003] Cardiovascular disease is a disease that seriously threatens human health, and early identification of individuals at high risk of cardiovascular disease is t...

AI Technical Summary

Problems solved by technology

If the arteries and veins segmented using the traditional segmentation network (for example, the output layer shares parameters and uses a multi-category softmax activation function), there are topological discontinuities, and then the key points cannot be identified based on the discontinuous topological map. Refined detection makes the measurement result single and lacks practicability

Images