Adaptive network suitable for high-reflection bright spot segmentation in retinal optical coherence tomography image

Abstract

The invention discloses a self-adaptive network suitable for high-reflection bright spot segmentation in a retinal optical coherence tomography image. The self-adaptive network comprises a feature coding module, a self-adaptive SA module and a feature decoding module, the feature coding module comprises a feature extraction unit and a dual residual DR module, the dual residual DR module comprisestwo residual blocks, the adaptive SA module comprises a feature input end, a deformable convolution layer, matrix multiplication and pixel-level summation, and the feature decoding module reconstructshigh-level features generated by the adaptive SA module, gradually performs feature splicing with local information guided by the dual residual DR module through 2 * 2 deconvolution layer deconvolution, and uses a result obtained through 1 * 1 convolution layer convolution as output of the feature decoding module. The method can simplify the learning process of the whole network and enhance gradient propagation while enhancing feature extraction, and can adapt to segmentation targets of different sizes.

Description

technical field [0001] The present application relates to the technical field of retinal OCT image segmentation, in particular to an adaptive network suitable for segmenting highly reflective bright spots in retinal optical coherence tomography images. Background technique [0002] Hard exudates, a prominent fundus change in diabetic retinopathy, appear as hyperreflective bright spots on optical coherence tomography imaging. In recent years, there have been many studies on the detection of hard exudation in retinal color photos, such as the detection method of hard exudation based on support vector machine, the automatic detection method based on k-nearest neighbor map region merging, the detection method based on threshold method, etc. A similar study also segmented bright spots in polarization-sensitive optical coherence tomography (Polarization-Sensitive Optical Coherence Tomography, PS-OCT). But segmentation methods based on deep learning are rare. At present, many dee...

AI Technical Summary

Problems solved by technology

However, the encoder-decoder structure of the original U-Net cannot realize the effective extraction and utilization of global features, and cannot adapt to segmentation objects of various shapes and sizes, and has poor segmentation performance for small objects.

Images