Automated image-based biomarker computation tools for diabetic retinopathy

描述（由申请人提供）：在这个STTR项目中，我们提出了EyeMark，一套使用视网膜图像照片自动计算糖尿病视网膜病变生物标志物的工具。具体来说，我们将开发计算微动脉瘤（MA）出现和消失率（联合称为周转率）的工具，作为监测糖尿病视网膜病变（DR）进展的生物标志物。可靠的基于图像的生物标志物的可用性将对DR护理的各个方面产生高度积极的影响，包括筛查、监测进展、药物发现和临床研究。有大量已发表的证据表明，MA周转率可以很好地预测进展为更严重视网膜病变的可能性，从而将MA周转率确立为糖尿病视网膜病变的优秀生物标志物。测量这一数量包括两个步骤：仔细校准当前和基线图像，以及标记单个MAs。如果完全由人工评分，这个过程非常耗时，而且容易出错。本项目的主要目标是通过自动化MA周转率测量所涉及的两个步骤来克服上述限制：准确的图像配准和MA检测。我们将开发端到端的桌面软件，用于自动计算MA周转量，并为临床医生提供直观的可视化工具，以更有效地监测糖尿病视网膜病变的进展。

DESCRIPTION (provided by applicant): In this STTR project, we present EyeMark, a set of tools for automated computation of biomarkers for diabetic retinopathy using retinal image photographs. Specifically, we will develop tools for computation of microaneurysm (MA) appearance and disappearance rates (jointly known as turnover rates) for use as a biomarker in monitoring progression of diabetic retinopathy (DR). The availability of a reliable image-based biomarker will have high positive influence on various aspects of DR care, including screening, monitoring progression, drug discovery and clinical research. There is ample published evidence that MA turnover rates are a good predictor of likelihood of progression to more severe retinopathy, establishing MA turnover as an excellent biomarker for diabetic retinopathy. Measuring this quantity involves two steps: careful alignment of current and baseline images, and marking of individual MAs. This process is very time consuming and prone to error, if done by entirely by human graders. The primary goal of this project is to overcome the above limitations by automating both the steps involved in MA turnover measurement: accurate image registration, and MA detection. We will develop end-too-end desktop software for automated computation of MA turnover and also provide intuitive visualization tools for clinicians to more effectively monitor diabetic retinopathy progression. PUBLIC HEALTH RELEVANCE: The proposed tool will greatly enhance the clinical care available to diabetic retinopathy patients by providing an automated tool for computation of a biomarker in a non-invasive manner. This will enable identification of patients who are more likely to progress to severe retinopathy, thus helping prevent vision loss in such patients by timely intervention. Early identification is especially important in face of long backlog of diabetic patients waiting for an eye examination, and the fact that 90% of vision loss can be saved by early identification. The 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.