Automated Screening for Diabetic Retinopathy by Content*

University Of Tennessee Health Sci Ctr, Memphis TN

Investigators

Linked publications & trials

Abstract

DESCRIPTION (provided by applicant): The application of computer-based imaging to the diagnosis of retinal disease is rapidly becoming a reality. Advances in the imaging of ocular anatomy and pathology can now provide data to diagnose and quantify specific diseases including diabetic retinopathy. The potential for these digital technologies is clear, new computer based systems, and diagnostic algorithms hold the promise of producing low-cost, potentially automated, diagnostic imaging systems. The goal of this project is to investigate the feasibility of a content-based image retrieval (CBIR) method to accurately describe and index human retinal images of diabetic retinopathy collected from low-cost, nondilated retinal photographic examinations. Our goal is to demonstrate the feature-based indexing and retrieval process of CBIR and verify our hypothesis, and novel concept, that retinal pathology can be identified and quantified from visually similar retinal images assembled from a large database comprising images of diabetic retinopathy. The proposed research extends this fundamental investigation by incorporating intrinsic and extrinsic patient data to provide a diagnostic method. We have brought together a unique team that is currently designing the algorithms, developing the analytical tools, and performing the required clinical trials to reach the stated goals of this RFA.