Automated image-based biomarker computation tools for diabetic retinopathy

AbstractIn this SBIR projectwe present EyeMarka set of advanced image analysis tools for automated computation of biomarkers for diabetic retinopathyDRusing retinal fundus imagesSpecificallywe will develop tools for computation of microaneurysmMAappearance and disappearance ratesjointly known as turnover ratesfor use as a biomarker in quantifying DR progression risk along with longitudinal analysis of other DR lesionsThe availability of a reliable image based biomarker will have high positive influence on various aspects of DR careincluding screeningmonitoring progressiondrug discovery and clinical researchMeasuring MA turnover and longitudinal analysis of DR lesions involves two labor intensive stepscareful alignment of current and baseline imagesand marking of individual lesionsThis process is very time consuming and prone to errorif done entirely by human gradersThe primary goal of this project is to overcome these limitations by automating both the steps involved in longitudinal analysisaccurate image registrationand lesion identificationWe have designed and developed a MA turnover computation prototype tool that robustly registers longitudinal imageseven with multiple lesion changesand effectively detects DR lesionslesion level AUROCandgtThe tool provides graceful degradation to confounding image factors by reporting MA turnover as a rangethereby capturing the inherent confidence in MA detectionBy the end of Phase IIB we will develop a market readyclinically validated end to end desktop software for robustautomated longitudinal lesion analysis and characterization that can work on the cloud to produce results in near constant timefor large datasetsand also provide intuitive visualization tools for clinicians to more effectively monitor DR progression Narrative The proposed toolEyeMarkwill greatly enhance the clinical care available to diabetic retinopathyDRpatients by providing an automated tool for computation of an imagebasedreliableDR biomarker in a non invasive mannerThis will enable identification of patients who are at higher risk to progress to severe retinopathythus helping prevent vision loss in such patients by timely interventionEarly identification is especially important in face of long backlog of diabetic patients waiting for an eye examinationand the fact thatof vision loss can be saved by early identificationThe availability of an effective biomarker will also positively influence the drug discovery process by facilitating early and reliable determination of biological efficacy of potential new therapies