Fundus image recognition model training method, fundus image recognition method and device

Abstract

The present invention provides a fundus image recognition model training method, a fundus image recognition method and equipment, wherein the fundus image recognition model training method includes: acquiring training data, the training data includes a first fundus image sample, a second fundus image sample And a third fundus image sample, wherein, the second fundus image sample and the first fundus image sample are fundus images of the same eye; the third fundus image sample and the first fundus image sample are different fundus images; The fundus image recognition model recognizes the first fundus image sample, the second fundus image sample and the third fundus image sample to obtain a loss value; adjusts parameters of the fundus image recognition model according to the loss value.

Description

technical field [0001] The invention relates to the technical field of medical image recognition, in particular to a fundus image recognition model training method, a fundus image recognition method and equipment. Background technique [0002] The current fundus diseases usually use special camera equipment to take fundus images. Doctors can judge whether the examinee may suffer from certain fundus diseases by observing the fundus images, so as to make suggestions on whether further examination or medical treatment is required. [0003] The condition of the fundus disease may continue to develop. In the follow-up process of the follow-up patients, the doctor needs to compare the previous fundus images to track the condition in order to give better treatment suggestions. In this way, it is necessary to choose among many fundus images Fundus images from the same eye, although doctors with many years of experience can select fundus images belonging to the same eye based on thei...

AI Technical Summary

Problems solved by technology

[0003] The condition of the fundus disease may continue to develop. In the follow-up process of the follow-up patients, the doctor needs to compare the previous fundus images to track the condition in order to give better treatment suggestions. In this way, it is necessary to choose among many fundus images Fundus images from the same eye, although doctors with many years of experience can select fundus images belonging to the same eye based on their own experience, but because there are many uncertain factors in the fundus shooting process, such as the brightness of the image, image rotation, translation, etc.

These will make it very difficult to identify the fundus image, and it will be difficult for doctors to accurately distinguish the fundus image from the same eye, making it difficult to accurately track the fundus condition

Images