Abstract
Choroid neovascularization (CNV) is one of the blinding ophthalmologic diseases. It is mainly caused by new blood vessels growing in choroid and penetrating Bruch's membrane. Accurate segmentation of CNV is essential for ophthalmologists to analyze the condition of the patient and specify treatment plan. Although many deep learning-based methods have achieved promising results in many medical image segmentation tasks, CNV segmentation in retinal optical coherence tomography (OCT) images is still very challenging as the blur boundary of CNV, large morphological differences, speckle noise, and other similar diseases interference. In addition, the lack of pixel-level annotation data is also one of the factors that affect the further improvement of CNV segmentation accuracy. To improve the accuracy of CNV segmentation, a novel multi-scale information fusion network (MF-Net) based on U-Shape architecture is proposed for CNV segmentation in retinal OCT images. A novel multi-scale adaptive-aware deformation module (MAD) is designed and inserted into the top of the encoder path, aiming at guiding the model to focus on multi-scale deformation of the targets, and aggregates the contextual information. Meanwhile, to improve the ability of the network to learn to supplement low-level local high-resolution semantic information to high-level feature maps, a novel semantics-details aggregation module (SDA) between encoder and decoder is proposed. In addition, to leverage unlabeled data to further improve the CNV segmentation, a semi-supervised version of MF-Net is designed based on pseudo-label data augmentation strategy, which can leverage unlabeled data to further improve CNV segmentation accuracy. Finally, comprehensive experiments are conducted to validate the performance of the proposed MF-Net and SemiMF-Net. The experiment results show that both proposed MF-Net and SemiMF-Net outperforms other state-of-the-art algorithms.
Highlights
Choroidal neovascularization (CNV), known as subretinal neovascularization, is a basic pathological change of various intraocular diseases such as age-related macular degeneration, central exudative chorioretinopathy, idiopathic choroidal neovascularization, pathological myopic macular degeneration, and ocular histoplasmosis syndrome (DeWan et al, 2006; Abdelmoula et al, 2013; Jia et al, 2014; Liu et al, 2015; Zhu et al, 2017)
In order to accurately segment Choroid neovascularization (CNV) and evaluate the performance of the proposed method, experienced ophthalmologists annotate pixel-level ground truth for the 1,522 optical coherence tomography (OCT) images with CNV collected from the UCSD public dataset (Kermany et al, 2018), which collected by the Shiley Eye Institute of the University of California San Diego (UCSD) and all of the images (Spectralis OCT, Heidelberg Engineering, Germany) were selected from retrospective cohorts of adult patients without exclusion criteria based on age, gender, or race
CE-Net achieves an increase of 0.21% for the main evaluation metric dice similarity coefficients (DSC), due to the combination of dense atrous convolution (DAC) and residual multi-kernel pooling (RMP)
Summary
Choroidal neovascularization (CNV), known as subretinal neovascularization, is a basic pathological change of various intraocular diseases such as age-related macular degeneration, central exudative chorioretinopathy, idiopathic choroidal neovascularization, pathological myopic macular degeneration, and ocular histoplasmosis syndrome (DeWan et al, 2006; Abdelmoula et al, 2013; Jia et al, 2014; Liu et al, 2015; Zhu et al, 2017) Multiscale global pyramid feature aggregation module and multi-scale adaptive-aware deformation module are proposed to segment corneal ulcer in slit-lamp image in our previous study (Wang et al, 2021b). To tackle these challenges and improve the CNV segmentation accuracy, a novel multi-scale information fusion network (MF-Net) is proposed for CNV segmentation in retinal OCT images. The experimental results show that, compared to state-of-the-art CNNbased methods, the proposed MF-Net achieves higher segmentation accuracy
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
