Retina anomaly analysis method and equipment

Abstract

The invention provides a retina anomaly analysis method and equipment. The method comprises the steps of conducting feature extraction on at least one fundus image through a feature extraction module; recognizing the extracted features through a region detection module, wherein the region detection module comprises a category branch and a regression branch, the category branch is used for obtaining region category information according to the features, and the regression branch is used for obtaining region position information according to the features; identifying the features through an overall anomaly detection module to obtain overall anomaly category information; segmenting a region corresponding to the region category information to obtain region area information; and carrying out statistical analysis on the region category information, the region area information, the region position information and the overall anomaly category information corresponding to the at least one fundus image to obtain anomaly correlation data which is used for reflecting a corresponding relationship between various overall anomaly categories and various region categories as well as the region area information and the region position information of the various region categories.

Description

technical field [0001] The invention relates to the field of medical data analysis, in particular to a retinal abnormality analysis method and equipment. Background technique [0002] Many human diseases such as macular degeneration, retinal tumors, diabetic retinopathy, arteriosclerosis, etc. will affect the eyes and cause changes in the retina. Because of its convenience and non-invasiveness, the fundus map can be used to detect abnormal changes of the fundus and monitor the progress of the disease. [0003] In recent years, machine learning has been widely used in the medical field, especially machine learning technology represented by deep learning has attracted widespread attention in the field of medical imaging. In terms of fundus image detection, deep learning technology has been used to detect fundus diseases such as glaucoma, diabetic retinopathy, and age-related macular degeneration, and achieved good results. However, in actual medical application scenarios, th...

AI Technical Summary

Problems solved by technology

However, in actual medical application scenarios, the diagnosis of glaucoma depends on the cup-to-disk ratio, the diagnosis of diabetes depends on hemorrhage, exudation, retinal neovascularization, drusen, geographic atrophy, and choroidal neovascularization on the basis of age-related macular degeneration. Retinal lesions such as blood vessels are not only closely related to the diagnosis of the disease, but also their location and number will also affect the severity of the fundus disease and the doctor's diagnosis plan

[0004] The existing technology has been able to detect and classify various retinal lesions using deep learning technology, but the existing detection results are difficult to reflect the severity of the disease or abnormality

Images