Edema region segmentation model based on OCT image

Abstract

The invention discloses an edema region segmentation algorithm based on an OCT (Optical Coherence Tomography) image, which comprises the following steps of: 1) image preprocessing: preprocessing an OCT image to be segmented by adopting Gaussian filtering; 2) coarsely segmenting an edema region: performing coarse segmentation on the OCT image by using a K-means algorithm to obtain a region of interest of the OCT image; and 3) carrying out fine segmentation on the edema region: taking the region of interest of the OCT image as an initial boundary, and carrying out fine segmentation on the OCT image based on an improved level set algorithm. The number of times of boundary line detection calculation iteration is reduced, and the segmentation efficiency is improved; the time efficiency is improved by about 23%, and the number of iterations is reduced by about 30%. In the aspect of segmentation precision, an SPF function is used for replacing an edge stop function to improve an existing level set model, the obtained model can converge to an edema area in the ROI, and therefore segmentation of the OCT image edema area is achieved.

Description

technical field [0001] The invention relates to the field of biomedical information and image processing, in particular to an edema region segmentation model based on OCT images. Background technique [0002] Diabetic macular edema (DME), caused by fluid leakage from damaged macular vessels and disruption of the retinal blood barrier, is characterized by intraretinal edema and retinal thickening, and is the primary cause of visual impairment in patients with diabetic retinopathy (DR) reason. [0003] Optical coherence tomography (Optical Coherence Tomography, OCT) is an image created by scanning based on Michelson interference principle. Regional site, which is widely used in the diagnosis of DME. Medical personnel usually use visual interpretation to interpret the edema area of ​​OCT to assess the severity of DME. The disadvantage of this method is that due to the influence of factors such as the experience of the doctor and the quality of the OCT image, the visual inter...

AI Technical Summary

Problems solved by technology

The disadvantage of this method is that, affected by factors such as the experience of the doctor and the quality of the OCT image, there is a large chance of visual interpretation of the edema area.

[0006] 1) Affected by noise, existing segmentation methods have over-segmentation when segmenting edema areas

[0007] 2) Affected by the ambiguity of the boundary, the existing segmentation methods have problems such as high computational complexity when searching for the boundary of the edema area

[0008] 3) Affected by the shape of the edema area, the existing segmentation methods have problems such as low accuracy and low calculation efficiency.

Images