Automatic Eye Disease Screening System

描述（由申请人提供）：该拟议项目的动机是，如果不引入视网膜图像的计算机辅助诊断，对年龄相关性黄斑变性（AMD）和糖尿病视网膜病变（DR）等眼部疾病进行全面、广泛的筛查在经济上是不可行的。根据CDC的数据，美国约有8000万人患有某种形式的眼病，其中包括2000万糖尿病患者有视网膜病变的风险，6000万青光眼的风险，1300万被诊断患有AMD。据估计，只有不到一半的糖尿病患者定期接受DR筛查，缺乏医疗保险和医疗保健提供者给近1000万糖尿病患者带来了重大障碍。创建一个负担得起的和可访问的解决方案，为这些糖尿病患者提供筛查服务，这对医疗保健界提出了一个重大挑战。美国公民利用眼科医生、验光师或其他受过训练的专家（“读者”）对每个病例进行分级的全面筛查计划将是极其昂贵的。解决办法是采用类似于乳房X光检查和子宫颈抹片检查等其他医疗应用的计算机辅助技术;这将为我们的高危人群提供全面、定期的筛查。这笔赠款将用于实施自动眼科疾病筛查系统（AEDSS），该系统将在美国和全球的中心进行高效筛查。这笔赠款将进行一项全面的多地点临床研究，以实施一个相互连接的区域系统，以评估基于计算机的算法的效率。这项研究将通过在德克萨斯州南部的一个阅读中心实施基础设施而成为可能，据估计，那里有近20万糖尿病患者，超过50%的人没有接受年度检查。该中心将是一个示范中心，可以记录软件的有效性，以便提交给FDA进行上市前批准（PMA）。其目标是表明中心的吞吐量，即每个时间段筛选的病例数，将通过使用混合（自动化-人类分级器）方法增加五倍，其灵敏度与单独基于人类的筛选相当或更好。这项拟议研究的意义是双重的。首先，通过自动化提高阅读中心的生产力，更多的高危糖尿病患者将获得这项服务，从而通过早期发现和治疗提高生产力和生活质量。其次，通过提供FDA批准的高效DR筛查系统，美国医疗保健系统将通过我们系统的效率提高而在经济上受益。我们的系统不会取代目前的人类读者;它只是允许增加五个或更多的因素的情况下，不牺牲灵敏度和特异性的吞吐量。筛查成本的降低将导致更多的阅读中心和高薪技术/医疗工作的增加。

DESCRIPTION (provided by applicant): This proposed project is motivated by the fact that comprehensive, broad-scale screening for eye diseases such as age-related macular degeneration (AMD) and diabetic retinopathy (DR) is economically prohibitive without the introduction of computer-assisted diagnosis of retinal images. According to the CDC approximately 80 million people in the U.S. have some form of eye disease, including 20 million diabetics at risk for retinopathy, 60 million at risk for glaucoma, and 13 million diagnosed with AMD. It is estimated that less than half of those individuals with diabetes are screened periodically for DR. Lack of medical coverage and access to healthcare providers imposes major obstacles for nearly 10 million diabetics. Creating an affordable and accessible solution to providing screening services to these diabetics presents a significant challenge to the healthcare community. A comprehensive screening program for U.S. citizens utilizing ophthalmologists, optometrists, or other trained specialists, "readers", to grade each case would be prohibitively expensive. The solution is to implement a computer-assisted technology similar to other medical applications, such as mammograms and Pap smears; that would provide comprehensive, periodic screening of our at-risk population. This grant will implement an automatic eye disease screening system (AEDSS) which will enable highly efficient screening at centers across the US and globally. This grant will conduct a comprehensive multi-site clinical study to implement a inter-connected regional system to evaluate the efficiency of computer-based algorithms. The study will be made possible through the implementation of the infrastructure for a reading center in South Texas where it is estimated that nearly 200,000 diabetics reside and over 50% do not receive annual examinations. The center will be a demonstration site where the efficacy of the software can be documented for submission to the FDA for pre-market approval (PMA). The goal is to show that a center's throughput, i.e. number of cases screened per time period, will increase five fold through the use of a hybrid (automation-human graders) approach with sensitivity that is comparable or better than solely human-based screening. The significance of this proposed research is two-fold. First, by increasing the productivity of reading centers through automation, a much larger population of at-risk diabetics will have access to this service, leading to improved productivity and quality of life through early detection and treatment. Second, by providing a FDA approved system for highly effective DR screening, the US healthcare system would benefit economically through the added efficiency of our system. Our system does not replace current human readers; it simply allows an increase of throughput of cases by factors of five or more without sacrificing sensitivity and specificity. The reduced cost of screening would lead to more reading centers and increased high-paying technical/medical jobs.