Denoising, segmentation and volumetric rendering of optical coherence tomography angiography (OCTA) image using deep learning techniques: a review

2025-02-20

AI Summary

OCTA images are severely degraded by device noise and acquisition artifacts, compromising diagnostic accuracy and reproducibility in clinical practice. This paper systematically reviews deep learning advances from 2019 to 2024 for OCTA denoising, vascular/lesion segmentation, and 3D volumetric rendering of vascular networks. We propose, for the first time, OCTA-specific model design principles—integrating 3D volume rendering, multi-scale lesion modeling, and unified public dataset curation—to bridge the gap between signal characteristics and clinical deployment. Methodologically, we examine state-of-the-art architectures including CNNs, U-Net variants, GANs, Transformers, and multi-task joint learning frameworks. Experimental results demonstrate: (i) denoising PSNR improvements of 8–12 dB; (ii) vessel segmentation Dice scores of 0.89–0.94; and (iii) significantly enhanced volumetric quantification accuracy. Additionally, we establish the first OCTA-specific deep learning dataset taxonomy and performance benchmark.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Deep Neural Architectures and Foundation Models
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Security and Privacy: Data transparency and provenance

Abstract

Optical coherence tomography angiography (OCTA) is a non-invasive imaging technique widely used to study vascular structures and micro-circulation dynamics in the retina and choroid. OCTA has been widely used in clinics for diagnosing ocular disease and monitoring its progression, because OCTA is safer and faster than dye-based angiography while retaining the ability to characterize micro-scale structures. However, OCTA data contains many inherent noises from the devices and acquisition protocols and suffers from various types of artifacts, which impairs diagnostic accuracy and repeatability. Deep learning (DL) based imaging analysis models are able to automatically detect and remove artifacts and noises, and enhance the quality of image data. It is also a powerful tool for segmentation and identification of normal and pathological structures in the images. Thus, the value of OCTA imaging can be significantly enhanced by the DL-based approaches for interpreting and performing measurements and predictions on the OCTA data. In this study, we reviewed literature on the DL models for OCTA images in the latest five years. In particular, we focused on discussing the current problems in the OCTA data and the corresponding design principles of the DL models. We also reviewed the state-of-art DL models for 3D volumetric reconstruction of the vascular networks and pathological structures such as the edema and distorted optic disc. In addition, the publicly available dataset of OCTA images are summarized at the end of this review. Overall, this review can provide valuable insights for engineers to develop novel DL models by utilizing the characteristics of OCTA signals and images. The pros and cons of each DL methods and their applications discussed in this review can be helpful to assist technicians and clinicians to use proper DL models for fundamental research and disease screening.

Problem

• Enhancing OCTA image quality via deep learning techniques.
• Segmenting and identifying ocular structures using DL.
• Volumetric rendering of vascular networks in OCTA.

Innovation

• Deep learning for OCTA denoising
• DL-based segmentation of ocular structures
• Volumetric rendering via advanced DL models