Optical coherence tomography angiography for vessel segmentation

• Main Author: Zhang, Mengyuan
• Other Authors: Liu Linbo
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2023
• Subjects: Engineering::Electrical and electronic engineering
• Online Access: https://hdl.handle.net/10356/167445
• Institution: Nanyang Technological University

Optical Coherence Tomography Angiography (OCTA) is a non-invasive imaging technique that has gained prominence in the diagnosis and management of various ocular diseases. Accurate vessel segmentation in OCTA images is crucial for the quantitative analysis of vascular structures. This study proposes a novel approach for OCTA vessel segmentation using Hessian-based filtering and local mean enhancement. The method involves pre-processing the images, extracting the Foveal Avascular Zone (FAZ) region, and applying a Hessian filter to enhance blood vessels. Local mean computation and thresholding techniques are employed for binary vessel segmentation and further enhancement. The proposed method was evaluated on a dataset of ten OCTA images, demonstrating promising results in vessel segmentation with improved accuracy and efficiency compared to existing techniques. This approach has potential applications in clinical practice and research, contributing to the better understanding and management of ocular diseases.