Fundus OCT image recognition method and device

Abstract

The invention provides a fundus OCT image recognition method and device, and the method comprises the steps: carrying out the recognition of a fundus OCT image through a neural network, and obtaininga recognition result related to lesion and fovea centralis; determining a region of interest according to the identification result; and determining the lesion degree according to the position and / orsize of the region of interest.

Description

technical field [0001] The invention relates to the field of ophthalmology image detection, in particular to a fundus OCT image recognition method and equipment. Background technique [0002] Machine learning technology has been widely used in the medical field, especially machine learning technology represented by deep learning has been widely concerned in the field of medical imaging. In terms of fundus image detection and differential diagnosis of eye diseases, deep learning technology has been used to detect single diseases such as glaucoma and diabetic retinopathy, and achieved good results. [0003] There are many types and large numbers of diseases that can be indicated by eye pictures, and the manifestations of the lesions are very different, ranging from large-scale fundus hemorrhage to very small hemorrhage and exudation. The diseases that can be indicated by fundus images include retinopathy (including diabetic retinopathy and other diseases), arteriosclerosis / hy...

AI Technical Summary

Problems solved by technology

The existing technology can use the neural network to classify OCT images to obtain the identification result of a certain disease. This scheme has certain accuracy, but the clinical application value of this identification result is small and the practicability is poor.

Images