Recent Patents on Computer Science

Hamid Mcheick  
Computer Science Department
University of Quebec at Chicoutimi
Chicoutimi, Quebec
Canada

Back

Retinal Vascular Features for Cardio Vascular Disease Prediction: Review

Author(s): Alauddin Bhuiyan, Ryo Kawasaki, Ecosse Lamoureux, Tien Y. Wong and Kotagiri Ramamohanarao

Affiliation: Eye Research Australia, University of Melbourne, Royal Victorian Eye and Ear Hospital, and Department of Computer Science and Software Engineering, University of Melbourne, VIC 3010, Australia.

Keywords: Cardiovascular Disease (CVD), optic disc, retinal fundus image, vascular features, vessel segmentation, vessel caliber, tortuosity, retinal imaging, hypertension, microaneurysms, hemorrhages, retinopathy, retinal neovascularisation, dyslipidemia, Retinal vascular features, vessel tortuosity, Retinal vessel caliber, Focal Arteriolar Narrowing (FAN), Arteriovenous Nicking (AVN), hemodynamic, microvasculature, cup-to-disc ratio, glaucoma, Principal Component Analysis, elliptical form, fovea, gradient vector flow, Gaussian filters, multi-threshold probing, curvilinear structure, Matched Filter Response, vector, k-nearest neighbor, Ribbon of Twins, topology, inner diameter, Framingham Risk Score, Prospective Cardiovascular Münster, Biomarker, Atherosclerosis Risk in Communities, Support Vector Machine

Abstract:

Recent advances in retinal imaging modality have enabled the identification of retinal vascular features which have been shown to predict cardiovascular diseases (CVD). Studies have shown that a number of retinal vascular features are associated with the signs of pre-clinical or clinical CVD. In this paper, we discuss these retinal vascular features and their association with CVD. We review the recent patents on feature extraction algorithms from retinal imaging. We explore existing algorithms on retinal vascular feature extraction; how these features are detected and analyzed from retinal images. We discuss the existing CVD prediction models and the potential for retinal vascular features which can improve the prediction models. We also illustrate the limitations of current retinal vascular feature analysis approaches in CVD prediction. Finally, we outline a future direction on retinal vascular feature analysis approach which can provide us more precise CVD prediction model.

Order Reprints Order Eprints Rights & PermissionsPrintExport

Article Details

VOLUME: 3
ISSUE: 3
Page: [164 - 175]
Pages: 12
DOI: 10.2174/2213275911003030164