Optical fundus image segmentation method and device, computer equipment and storage medium

Abstract

The invention discloses an optical fundus image segmentation method and device, computer equipment and a storage medium, and the method comprises the steps: employing an optical coherence tomography technology to collect an original optical fundus image, carrying out the image reconstruction of the original optical fundus image, and obtaining a high-resolution optical fundus image; performing dataamplification on the high-resolution optical fundus image, and performing network training by using the high-resolution optical fundus image subjected to data amplification processing to obtain a choroid region segmentation network model; and inputting the collected to-be-processed optical fundus image into the choroid region segmentation network model, and segmenting a choroid region from the to-be-processed optical fundus image. By constructing the choroid segmentation network model, the choroid region can be automatically segmented from the optical fundus image, and the thickness of the segmented choroid can be automatically calculated, so that the image noise in the optical fundus image can be eliminated, the measurement precision of the choroid can be improved, and the subjective deviation can be eliminated.

Description

technical field [0001] The invention relates to the technical field of optical image processing, in particular to an optical fundus image segmentation method, device, computer equipment and storage medium. Background technique [0002] The imaging quality of optical images plays an important role in medical analysis. Optical coherence tomography (OCT) is a fast-growing and most promising new tomography technology in recent years. It has the characteristics of non-contact, non-damage and real-time measurement. Non-contact, non-invasive tomographic imaging of structures. OCT is an optical analog of ultrasound. Its axial resolution depends on the coherence characteristics of the light source, up to 10um, and the penetration depth is almost not limited by the transparent refractive medium of the eye. It can observe the anterior segment and display the posterior segment The morphological structure has a wide range of applications in the acquisition of eye images. [0003] In r...

AI Technical Summary

Problems solved by technology

[0006] The technical problem to be solved by the present invention is to provide an optical fundus image segmentation method, device, computer equipment and storage medium for the above-mentioned defects of the prior art, aiming to solve the problem that the choroidal thickness measurement method in the prior art mainly relies on Manual measurement of OCT fundus images leads to problems such as large image noise, limited resolution in the visual axis direction, few measurement points, subjective deviation, and low measurement accuracy.

Images