A level set and deep learning combined retinal layer segmentation method and system

Abstract

The invention discloses a level set and deep learning combined retina layer segmentation method and system, and the method comprises the steps: carrying out the preprocessing of an image collected byoptical coherence tomography equipment, enabling an active contour to evolve to a to-be-segmented boundary through a level set model through employing the priori information of the thickness of a retina layer, collecting normal retina OCT images and pathological retina OCT images, making classification labels, and constructing a training set and a test set; training a deep learning network by using the image data in the training set; processing the image data in the test set by using the trained deep learning network to obtain a retinal layer coarse classification result; and taking a retina layer coarse classification result as a level set function, evolving each level set function by adopting a gradient descent method, and obtaining a fine retina layer boundary.

Description

technical field [0001] The present disclosure relates to the field of image segmentation, specifically an OCT retinal layer segmentation method and system based on a combination of a level set model and an FCN deep learning network. Background technique [0002] Optical coherence tomography (OCT) provides a noninvasive, noncontact imaging technique that can be used to generate high spatial resolution images of retinal structures. Accurate quantification of retinal structure can be a very tedious task due to the high precision and large amount of data it requires, and a critical step in the quantification of retinal layer thickness is segmentation. [0003] At present, the OCT retinal layer boundary segmentation methods mainly include: graph search algorithm, active contour and level set method and machine learning method. The graph search algorithm uses dynamic programming technology to locate the retinal layers of normal people and age-related macular degeneration patients...

AI Technical Summary

Problems solved by technology

During the research and development process, the inventors found that although deep learning and machine learning methods perform well in hierarchical segmentation, they require a large number of labeled samples

Images