Research Project
9622980
Abstract In this SBIR project, we present EyeMark, a set of advanced image analysis tools for automated computation of biomarkers for diabetic retinopathy (DR) using retinal fundus images. Specifically, we will develop tools for computation of microaneurysm (MA) ap- pearance and disappearance rates (jointly known as turnover rates) for use as a bi- omarker in quantifying DR progression risk along with longitudinal analysis of other DR lesions. The availability of a reliable image-based biomarker will have high positive influ- ence on various aspects of DR care, including screening, monitoring progression, drug discovery and clinical research. Measuring MA turnover and longitudinal analysis of DR lesions involves two labor in- tensive steps: careful alignment of current and baseline images, and marking of individual lesions. This process is very time consuming and prone to error, if done entirely by human graders. The primary goal of this project is to overcome these limitations by automating both the steps involved in longitudinal analysis: accurate image registration, and lesion identification. We have designed and developed a MA turnover computation prototype tool that ro- bustly registers longitudinal images (even with multiple lesion changes) and effectively detects DR lesions (lesion level AUROC>=0.95). The tool provides graceful degradation to confounding image factors by reporting MA turnover as a range, thereby capturing the inherent confidence in MA detection. By the end of Phase IIB we will develop a market ready, clinically validated end-to-end desktop software for robust, automated longitudinal lesion analysis and characterization that can work on the cloud to produce results in near constant time (for large datasets), and also provide intuitive visualization tools for clinicians to more effectively monitor DR progression.
