Mobile App for Diabetic Retinopathy Screening using Cellphone Retinal Camera

使用手机视网膜摄像头进行糖尿病视网膜病变筛查的移动应用程序

• 批准号: 8782362
• 负责人: Kaushal Solanki
• 金额: $ 21.7万
• 依托单位: EYENUK, INC.
• 依托单位国家: 美国
• 财政年份: 2014
• 资助国家: 美国
• 起止时间: 2014-09-01 至 2015-12-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/8782362
• 关键词: Address; Adult; Age; Agreement; Algorithms; Architecture; Area; Back; Biomedical Engineering; Blindness; Chairperson; Color; Computer Vision Systems; Computer software; Descriptor; Detection; Devices; Diabetic Retinopathy; Diagnosis; Diagnostic; Elements; Ensure; Equipment; Eye; Foundations; Friends; Fundus; Goals; Gold; Home environment; Human; Image; Image Analysis; Leadership; Lesion; Machine Learning; Marketing; Measures; Monitor; Nurses; Ophthalmology; Optics; Outcome; Output; Patients; Pattern Recognition; Phase; Population; Process; Public Health; ROC Curve; Recommendation; Research; Resolution; Retinal; Retinal Diseases; Sensitivity and Specificity; Small Business Innovation Research Grant; Small Business Technology Transfer Research; Software Engineering; Software Tools; Solid; Specialist; Spouses; Staging; Surveys; System; Technology; Telephone; Testing; Time; Training; Underserved Population; Universities; Washington; Work; base; clinical care; cloud based; cluster computing; commercialization; computerized; cost; design; diabetic; diabetic patient; experience; glucose monitor; image processing; improved; interest; longitudinal analysis; member; mobile application; novel; point of care; prevent; professor; prototype; public health relevance; research and development; screening; statistics; success; tool

DESCRIPTION (provided by applicant): In this SBIR project, we present EyeApp, a smartphone-based end-to-end point-of- care diabetic retinopathy diagnostic device comprising of a smartphone-attachable retinal imaging camera and validated software application enabling fully-automated screening of diabetic retinopathy (DR). DR is the leading cause of new-onset blindness in working-age adults in the industrialized world today. Studies show that in 90% of the cases vision loss can be prevented if DR is diagnosed at an early stage. Recommended annual screening is expected to significantly improve outcome for diabetic patients, but even in the developed world, majority of the diabetics do not regularly have the annual screening due to several limiting factors. To make things worse, the current cost of screening all diabetics to detect those with early DR is very high because: (a) most patients will have normal exams (thus, causing inefficient use of specialist's time), (b) retinal imaging equipment is very expensive, and (c) trained technicians must operate the imaging equipment. In this project we address these critical limitations by employing cellphone retinal camera attachment, the Ocular CellScope, developed by our collaborators at UCB/UCSF, which simply attaches to an iPhone, doesn't need a trained technician to operate, will obtain wide-field images, and will cost several orders of mag- nitude less than the current tabletop retinal imaging cameras. Diabetic patients can attach the camera to their smartphones and with some assistance (untrained friend, spouse, or nurse) obtain retinal video for each eye using an app residing on the same phone. This app guides the image capture to ensure that best quality frames are obtained, and also ensure all areas of the eye are imaged enabling wide-field imaging. EyeApp's validated algorithms output a near-instant DR screening recommendation score. EyeApp is con- ceptualized as a culmination of over three years of research and development at Eyenuk (on computerized DR screening) and at UCB/UCSF (on smartphone retinal camera) that has already produced functional prototypes of critical technology modules. This phase I project focusses on seamless integration of these proven modules (the camera hardware and the diagnostic software) via several novel ideas.

描述（由申请人提供）：在该SBIR项目中，我们提出了EyeApp，一种基于智能手机的端到端护理点糖尿病视网膜病变诊断设备，包括智能手机可连接的视网膜成像相机和经验证的软件应用程序，能够实现糖尿病视网膜病变（DR）的全自动筛查。DR是当今工业化国家工作年龄成人新发失明的主要原因。研究表明，在90%的病例中，如果在早期阶段诊断出DR，视力丧失是可以预防的。 推荐的年度筛查预计将显着改善糖尿病患者的预后，但即使在发达国家，由于几个限制因素，大多数糖尿病患者也没有定期进行年度筛查。更糟糕的是，目前筛查所有糖尿病患者以检测早期DR患者的成本非常高，因为：（a）大多数患者将进行正常检查（因此，导致专家时间的低效利用），（B）视网膜成像设备非常昂贵， (c)必须由经过培训的技术人员操作成像设备。在这个项目中，我们通过使用手机视网膜相机附件来解决这些关键限制，Ocular CellScope由我们在UCB/UCSF的合作者开发，它简单地连接到iPhone上，不需要训练有素的技术人员来操作，将获得宽视场图像，并且将比当前桌面视网膜成像相机少几个数量级。糖尿病患者可以将摄像头连接到他们的智能手机上，并在一些帮助下（未经培训的朋友，配偶或护士）使用同一手机上的应用程序获得每只眼睛的视网膜视频。该应用程序指导图像捕获，以确保获得最佳质量的帧，并确保眼睛的所有区域都成像，从而实现宽视场成像。EyeApp经过验证的算法可以输出近乎即时的DR筛查推荐评分。EyeApp被认为是Eyenuk（计算机化DR筛查）和UCB/UCSF（智能手机视网膜摄像头）三年多研发的结晶，已经生产出关键技术模块的功能原型。第一阶段项目的重点是通过几个新颖的想法无缝集成这些成熟的模块（相机硬件和诊断软件）。