High-resolution retinal vascular imaging and analysis system

Che Azemin, Mohd Zulfaezal and Mohd Tamrin, Mohd Izzuddin and Yusof @ Alias, Firdaus and Salam, Adzura (2024) High-resolution retinal vascular imaging and analysis system. International Journal of Allied Health Sciences, 8 (Suppl. 3). p. 46. ISSN N/A E-ISSN 2600-8491

Preview

PDF (Abstract) - Published Version Download (477kB) | Preview

Abstract

PURPOSE: This study aims to develop and evaluate the High-Resolution Retinal Vascular Imaging and Analysis System (HRVIAS), an advanced diagnostic tool designed to improve the evaluation and monitoring of retinal vascular health. METHODS: HRVIAS utilizes high-resolution retinal images (2048x2048 pixels) and incorporates deep learning models specifically trained for vascular feature segmentation. The system employs adaptive thresholding methods and fractal dimension analysis to enhance image clarity and provide quantitative measures of retinal vascular complexity. ROC analysis is used to determine cut-off values for fractal dimension assessments, and the system is designed for both high-performance computing systems and more modest specifications to ensure broad accessibility. RESULTS: The HRVIAS demonstrated superior performance in segmenting and analyzing retinal vascular features compared to traditional methods. The adaptive thresholding technique provided significant improvements in the visibility of vascular structures, and fractal dimension analysis yielded precise quantifications of vascular complexity. ROC curve analysis indicated high diagnostic accuracy for various ocular diseases, including Age-related Macular Degeneration (AMD), Diabetic Retinopathy (DR), and Glaucoma. CONCLUSION: The HRVIAS represents a significant advancement in ocular health diagnostics by leveraging cutting-edge technology to enhance diagnostic accuracy and early disease detection. Its ability to provide detailed risk assessments and improve the visibility of retinal structures makes it an invaluable tool for practitioners, ultimately leading to better patient outcomes.

Item Type:	Article (Abstract)
Additional Information:	6768/114283
Uncontrolled Keywords:	Retinal imaging, deep learning, vascular segmentation, adaptive thresholding, fractal dimension analysis
Subjects:	R Medicine > RE Ophthalmology
Kulliyyahs/Centres/Divisions/Institutes (Can select more than one option. Press CONTROL button):	Kulliyyah of Allied Health Sciences Kulliyyah of Allied Health Sciences > Department of Optometry and Visual Science Kulliyyah of Medicine Kulliyyah of Medicine > Department of Ophthalmology Kulliyyah of Information and Communication Technology Kulliyyah of Information and Communication Technology Kulliyyah of Information and Communication Technology > Department of Information System Kulliyyah of Information and Communication Technology > Department of Information System
Depositing User:	Dr. Mohd Zulfaezal Che Azemin
Date Deposited:	05 Sep 2024 16:13
Last Modified:	05 Sep 2024 16:21
URI:	http://irep.iium.edu.my/id/eprint/114283

Actions (login required)

View Item