Deep learning-based segmentation and quantitative analysis of retinal microstructures in optical coherence tomography angiography images using RS_Unet3+

2022/12/26 by Ke Ma, Xiao, Peng, Jinze Zhang +13 · 1 voice
Medicine · Computer Science · #Retinal Imaging and Analysis #Glaucoma and retinal disorders #Digital Imaging for Blood Diseases

paper · pdf · doi:10.1016/j.imed.2025.05.008

openalex publication_date 2025/09/17 · openalex created_date 2025/10/10 · openalex updated_date 2026/05/30

Abstract

Optical coherence tomography angiography (OCTA) is a novel, non-invasive imaging technique that enables capillary-level visualization of retinal vasculature, offering critical insights into various ophthalmic diseases. Accurate segmentation and quantitative analysis of microstructures—specifically the retinal vascular network (RVN) and foveal avascular zone (FAZ)—are essential for diagnosis and treatment planning. This study aims to develop an artificial intelligence-based system that automates the segmentation and analysis of OCTA microstructures using a deep learning framework. OCTA images were retrospectively collected from January 2020 to December 2022, comprising two public datasets (ROSE-1 and OCTA₃M) and a newly constructed clinical dataset, FAROS, acquired at Zhongshan Ophthalmic Center. The FAROS dataset included 40 en face OCTA images from 40 eyes (20 healthy and 20 with retinal diseases such as diabetic retinopathy, age-related macular degeneration, and retinal vein occlusion). In addition, a separate clinical dataset containing 20 eyes with diabetic retinopathy (DR) was enrolled to verify the clinical consistency of observed parameter trends. To accurately segment the RVN and FAZ, we firstly introduced an innovative OCTA microstructure segmentation network (RS_Unet3+) by combining an encoder-decoder-based architecture with full-scale skip connections and the split-attention-based residual network ResNeSt, paying specific attention to OCTA microstructural features while facilitating better model convergence and feature representations. We then performed multiclass segmentation on the FAROS dataset using proposed RS_Unet3+ and automatically calculated multiple RVN and FAZ parameters from the segmented image for quantitative analysis. Primary quantitative parameters included FAZ area (A), perimeter (P), circularity index (CI), vessel perfusion density (VPD), vessel length density (VLD), fractal dimension (FD), and tortuosity (T). Statistical comparisons between healthy and diseased groups were conducted using the Mann–Whitney U test with a significance threshold of P ≤ 0.05. The proposed RS_Unet3+ is verified through systematic experiments to achieve excellent single-task/multi-class performances for RVN or/and FAZ segmentation on two publicly available OCTA datasets, respectively. The multiclass segmentation task on the FAROS dataset using RS_Unet3+ was also achieved excellent segmentation performance that could be used as baseline performance as well. In the independent clinical DR dataset, P of FAZ were significantly higher in the DR group compared to healthy controls ( P <0.01), while the CI was significantly lower ( P < 0.001). Vessel parameters including VLD and FD were significantly reduced ( P < 0.05, P <0.01, respectively), and T was increased ( P < 0.05) in DR eyes, aligning with previously reported clinical patterns. The RS_Unet3+-based deep learning framework enables accurate segmentation and automated quantitative evaluation of retinal microstructures in OCTA images, which is expected to be a potentially reliable and convenient auxiliary tool for clinical disease diagnosis and treatment.

Citations

• CS2-Net: Deep Learning Segmentation of Curvilinear Structures in Medical Imaging
• Swin UNETR: Swin Transformers for Semantic Segmentation of Brain Tumors in MRI Images
• Medical SAM Adapter: Adapting Segment Anything Model for Medical Image Segmentation
• Retinal Structure Detection in OCTA Image via Voting-based Multi-task Learning
• Transformers in Medical Image Analysis: A Review
• D-Former: A U-shaped Dilated Transformer for 3D Medical Image Segmentation
• BSDA-Net: A Boundary Shape and Distance Aware Joint Learning Framework for Segmenting and Classifying OCTA Images
• ROSE: A Retinal OCT-Angiography Vessel Segmentation Dataset and New Model
• ResNeSt: Split-Attention Networks
• UNet 3+: A Full-Scale Connected UNet for Medical Image Segmentation
• UNet 3+: A Full-Scale Connected UNet for Medical Image Segmentation
• UNet++: Redesigning Skip Connections to Exploit Multiscale Features in Image Segmentation
• Focal Loss for Dense Object Detection
• Deep Residual Learning for Image Recognition
• U-Net: Convolutional Networks for Biomedical Image Segmentation
• Fully Convolutional Networks for Semantic Segmentation

Discussions

• Deep learning-based segmentation and quantitative analysis of retinal microstructures in optical coherence tomography angiography images using RS_Unet3+ doi.org/10.1016/j.im... [bsky, 0 points, 0 comments]