Computationally Efficient Artifact Suppression for Optical Coherence Tomography Angiography

In optical coherence tomography angiography (OCTA), vessels exhibit tail artifacts beneath them in B-scan images. These superficial vessel tails cast shadows on deeper vascular layers, significantly degrading image quality. Current artifact suppression methods often require complex algorithms and ex...

Full description

Saved in:
Bibliographic Details
Main Authors: Wenxin Zhang, Jun Pan, Shiqi Tang, Ning Liu
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/11126059/
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In optical coherence tomography angiography (OCTA), vessels exhibit tail artifacts beneath them in B-scan images. These superficial vessel tails cast shadows on deeper vascular layers, significantly degrading image quality. Current artifact suppression methods often require complex algorithms and extensive computation, limiting their practicality. To address this, we propose a computationally efficient image processing technique that simultaneously reduces artifacts to advance clinical OCTA applications. Our method integrates OCT structural images and OCTA images, and selectively enhances the blood-vessel wall interface, using vessel wall signals as markers for vascular positioning, achieving shorter tail artifacts with minimal processing overhead. Experiment results demonstrate that our approach requires only one-tenth of the computation time of traditional cmOCTA techniques, showing more significant speed improvements compared to other artifact suppression methods. Moreover, our method achieves effective artifact suppression comparable to conventional artifact removal methods at this high speed. Experimental validation using palm vasculature demonstrates the potential of our approach for real-time imaging and suggests that it may support clinical scenarios requiring timely feedback.
ISSN:2169-3536