Image recognition method based on OCT image, server and storage medium

Abstract

The invention relates to the field of artificial intelligence, the invention discloses an image recognition method based on an OCT image. The method is applied to a server, the method comprises the following steps: acquiring an OCT image without an abnormal region as a sample image to construct a generative adversarial network; respectively training a generator and a discriminator of the generative adversarial network to obtain a target discriminator and a target generator; alternately iterating the target generator and the target discriminator to train the generative adversarial network untilthe training is completed; and obtaining a to-be-identified image uploaded by the client, inputting the to-be-identified image into the trained generative adversarial network to obtain an analog image, calculating an abnormal score between the analog image and the to-be-identified image by using a first algorithm, and when the abnormal score is greater than a second preset threshold, judging thatthe to-be-identified image is an abnormal image containing an abnormal region. According to the invention, the accuracy of identifying and judging whether the information reflected in the OCT image is abnormal or not can be improved.

Description

technical field [0001] The present invention relates to the field of artificial intelligence, in particular to an image recognition method based on OCT images, a server and a storage medium. Background technique [0002] OCT (Optical Coherence Tomography, Optical Coherence Tomography) is an imaging technology that has developed rapidly in the past decade. It uses the basic principle of weakly coherent light interferometer to detect the back of the incident weakly coherent light at different depths of biological tissue. Reflected or several scattered signals can be scanned to obtain two-dimensional or three-dimensional structural images of biological tissues, that is, OCT images. Due to the particularity of OCT images, it is usually necessary to use specific instruments to identify whether the information reflected in the corresponding OCT images is abnormal. Not only the accuracy of image recognition is low, but also the efficiency of image result recognition is not high. Wi...

AI Technical Summary

Problems solved by technology

Due to the particularity of OCT images, it is usually necessary to use specific instruments to identify whether the information reflected in the corresponding OCT images is abnormal. Not only the accuracy of image recognition is low, but also the efficiency of image result recognition is not high. With the rapid development of neural networks, More and more neural networks are also being applied to intelligently identify whether OCT images are abnormal

[0003] Since most of the existing neural networks need to use a large number of samples of abnormal OCT images (that is, images containing suspected lesion areas) in the training process, in practice, due to the privacy issues of patients involved in abnormal OCT images, they cannot be compared with normal OCT images. It is so easy to obtain, which leads to many difficulties in the application of the existing neural network in the medical field. Even the recognition model obtained through a small amount of abnormal OCT image training has the problem of low recognition accuracy.

Images