Feature Extraction of Retinal Microvasculature of Retinal Images

MOHAMMED ENAMUL, HOQUE (2018) Feature Extraction of Retinal Microvasculature of Retinal Images. [Final Year Project Report] (Unpublished)

	PDF Feature Extraction of Retinal Microvasculature of Retinal24pgs.pdf Download (1MB)
	PDF (Please get the password by email to repository@unimas.my , or call ext: 082-583914/3973/3933) M ENAMUL HOQUE.pdf Restricted to Registered users only Download (3MB)

Abstract

Image Processing, more generally digital image processing is one of the most widely used computer vision technology, especially in Biomedical engineering. Modern ophthalmology is directly dependent on this robust technology, digital image processing to find out the biomarkers analyzing the fundus eye images that are responsible for different kinds of life-threatening diseases like diabetic retinopathy, macular degeneration, hypertensive retinopathy, transient ischemic attack or sharp stroke and some other cardiovascular disease. The geometric features like vessel tortuosity, branching angles, vessel diameter and fractal dimension are considered as the biomarkers for the cardiovascular diseases mentioned above. Retinal vessel diameter widening has found as the early symptom of transient ischemic attack or sharp stroke. A succinct and meaningful review of the latest quantitative diagnostic methods that are developed employing the digital image analysis principles for measuring the remarkable features mainly the vessel diameter has been provided in the literature of this project. In this project, a completely new and computer-aided automatic method to measure the retinal vessel diameter employing the Euclidean Distance Transform technique has been developed. The proposed system measures the Euclidean Distance of the bright pixels exist on the Region of Interest (ROI). Further, the proposed system was evaluated on the High-Resolution Fundus Image Database (HRFID) and Retinal Vessel Image set for Estimation of Width (REVIEW) Database. The HRFID was used to evaluate the performance of the segmentation technique that was employed in this project and obtained 94.3% accuracy with 66.5% Sensitivity, 97.86% Specificity, 77.265 Positive Predictive Value (PPV) and 96.60% Negative Predictive Value (NPV). The Vascular Disease Image Set (VDIS) and Central Light Reflex Image Set (CLRIS) of REVIEW database were used to evaluate the overall system performance that measures the vessel diameter. The proposed system obtained 98.1% accuracy for the CLRIS and 97.7% accuracy for VDIS. With further evaluation, validation and enhancement of the method, it can be integrated into the clinical computer-aided diagnostic tool. The methodology and the evaluation results are explained in this report.

Actions (For repository members only: login required)

View Item