GA lesion segmentation method based on depth-concatenated model for SD-OCT images

Abstract

The invention discloses a GA lesion segmentation method based on depth-concatenated model for SD-OCT images. The method comprises steps of three kinds of depth network models with different layers arefirstly configured; The first layer is the input layer, the last layer is the output layer, and the middle layer is a sparse self-encoder with different number of neurons, and the encoding and decoding process are symmetrically distributed. The training is divided into two stages, self-supervised feature extraction stage and supervised-based classifier training stage. After the first stage training is completed, the first stage coding process is added with soft-max loss function training basic classifier inputs the labeled positive and negative samples with h-dimension characteristics into the depth network model, and trains the output layer of the soft-max classifier obtains the final segmentation result. Finally, based on Adaboost cascade strategy, the training process of the above model is fused to improve the final segmentation results. This method improves the segmentation accuracy of GA lesions and is of great significance for the prevention and diagnosis of macular diseases inthe elderly.

Description

technical field [0001] The invention relates to a lesion segmentation method, in particular to a frequency-domain optical coherence tomography retinal image segmentation method for geographic atrophy lesions of a depth cascade model. Background technique [0002] Retinopathy is the main factor affecting vision loss, among which age-related macular degeneration (Age-related Macular Degeneration, AMD) has become one of the main diseases affecting the vision of the elderly. In recent years, due to the characteristics of fast imaging speed and high resolution, SD-OCT imaging technology has been widely used in the diagnosis and treatment of retinal diseases. It can effectively present changes in the grayscale and structure of retinal tissue layers. The results of clinical experiments show that SD-OCT retinal images can show a variety of retinopathy tissues caused by AMD, such as geographic atrophy, drusen, retinal detachment, edema, etc. Since GA lesions are often associated wit...

AI Technical Summary

Problems solved by technology

[0004] (2) Unsupervised segmentation method

[0008] However, in terms of sample selection, artificially randomly selecting a fixed number of positive and negative samples is likely to lose some of the original information of the data; in terms of feature expression, due to the complexity of the depth features of 3D SD-OCT images, only one deep network It is difficult for the mo...

Images