Advanced Image Analysis Tools for Diabetic Retinopathy Telemedicine Application

用于糖尿病视网膜病变远程医疗应用的高级图像分析工具

• 批准号: 9142236
• 负责人: Kaushal Solanki
• 金额: $ 98.4万
• 依托单位: EYENUK, INC.
• 依托单位国家: 美国
• 财政年份: 2016
• 资助国家: 美国
• 起止时间: 2016-08-01 至 2018-07-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/9142236
• 关键词: Adult; Affect; Age; Algorithms; Architecture; Area; Background Diabetic Retinopathy; Blindness; Caliber; Caring; Certification; Clinical; Clinical Research; Clinical Trials; Color; Communication; Computer Vision Systems; Consult; County; Data; Data Collection; Data Set; Detection; Diabetes Mellitus; Diabetic Angiopathies; Diabetic Retinopathy; Early Diagnosis; Early treatment; Effectiveness; Electronic Health Record; Europe; European; Exudate; Eye; Faculty; Fundus; Goals; Gold; Health; Health Services; Healthcare; Hour; Image; Image Analysis; Imagery; Industry; Information Systems; Institutes; Italy; Lesion; Los Angeles; Manufacturer Name; Marketing; Masks; Measures; Medical Device; Medicine; Observational Study; Ophthalmic examination and evaluation; Ophthalmologist; Ophthalmology; Optometry; Patients; Pattern Recognition; Phase; Pilot Projects; Population; Primary Health Care; Productivity; Protocols documentation; Provider; Race; Reading; Recommendation; Recruitment Activity; Reporting; Research; Resources; Retinal; Risk; Safety; Sensitivity and Specificity; Site; Small Business Innovation Research Grant; Small Business Technology Transfer Research; Specificity; Surrogate Markers; System; Telemedicine; Testing; Triage; Universities; Validation; Wisconsin; Work; adjudicate; base; bioimaging; clinical care; cloud based; computerized; cost effective; diabetic; diabetic patient; digital imaging; experience; high throughput analysis; image processing; improved; macular edema; programs; public health relevance; safety net; screening; software systems; success; tool

 DESCRIPTION (provided by applicant): In this Phase IIB application, we propose to commercialize and bring to market EyeArt, a fully automated, high throughput, retinal image analysis based diabetic retinopathy (DR) screening tool (which has been developed and already validated in bench tests on very large real-world data) by conducting formal clinical trials and integrating it into several clinical information systems. With its fully automated use, high throughput capability (can analyze several thousands of patient cases in just a few hours), scalable architecture, and ability to seamlessly integrate into clinical workflows, EyeArt will aid the expansion of DR screening and help bridge the exponentially growing disparity between the number of diabetic patients and the number of eye care providers. DR is a common microvascular complication of diabetes affecting 80% of all diabetic patients. Even though vision loss due to DR is preventable by early detection and treatment, it is the one of the leading global causes of preventable blindness primarily due to the lack of easy access to eye care providers. Automated DR screening is the only way to make DR screening more accessible to the large and growing population of diabetic patients, currently 29 million in the US and 387 million world-wide and expected to grow to 110 million and 592 million respectively by 2030. To help reduce risk of DR vision loss in the diabetic population, EyeArt uses advanced image analysis algorithms to make DR screening more efficient, cost-effective, and more accessible. EyeArt has been validated and shown to have high screening sensitivity (safety) and efficacy on multiple large real-world datasets including one with over 54,000 patient cases (and 434,000 images) obtained from EyePACS, a DR telescreening system with over 360 sites in the US and elsewhere. Eyenuk has received ISO 13485 certification as a medical device manufacturer and EyeArt has received CE Marking and is commercially available in Europe. Going forward, we will clinically validate EyeArt in formal clinical trials and integrate it into clinical informatio systems to enable EyeArt's seamless incorporation into existing clinical workflows. This will pave the way for EyeArt's availability to the US DR screening market and reduce the burden on clinical resources by improving health care productivity. This will help increase the DR screening conformance in the diabetic population and thus help in reducing and ultimately eliminating vision loss due to DR through early detection and treatment.

 描述（由申请人提供）：在本IIB期申请中，我们建议通过进行正式临床试验并将其集成到几个临床信息系统中，将EyeArt商业化并推向市场，EyeArt是一种全自动、高通量、基于视网膜图像分析的糖尿病视网膜病变（DR）筛查工具（已开发并已在非常大的真实世界数据的台架测试中验证）。凭借其全自动化的使用，高通量能力（可以在短短几个小时内分析数千例患者病例），可扩展的架构以及无缝集成到临床工作流程中的能力，EyeArt将有助于 DR筛查的扩大有助于弥合糖尿病患者数量与眼科护理提供者数量之间呈指数增长的差距。DR是糖尿病常见的微血管并发症，影响80%的糖尿病患者。尽管DR导致的视力丧失可以通过早期检测和治疗来预防，但它是全球领先的视力丧失之一 可预防的失明的主要原因是缺乏方便的眼科护理提供者。自动DR筛查是使DR筛查更容易为庞大且不断增长的糖尿病患者人群提供的唯一途径，目前美国有2900万糖尿病患者，全球有3.87亿糖尿病患者，预计到2030年将分别增长到1.1亿和5.92亿。 为了帮助降低糖尿病人群DR视力丧失的风险，EyeArt使用先进的图像分析算法，使DR筛查更有效，更具成本效益，更容易获得。EyeArt已经过验证，并在多个大型真实世界数据集上显示出高筛查灵敏度（安全性）和有效性，包括从EyePACS获得的超过54，000例患者病例（和434，000张图像），EyePACS是一种DR筛查系统，在美国和其他地方拥有超过360个站点。Eyenuk作为医疗设备制造商已获得ISO 13485认证，EyeArt已获得CE认证，并在欧洲上市。展望未来，我们将在正式的临床试验中对EyeArt进行临床验证，并将其整合到临床信息系统中，使EyeArt能够无缝融入现有的临床工作流程。这将为EyeArt进入美国DR筛查市场铺平道路，并通过提高医疗保健生产力来减轻临床资源的负担。这将有助于提高糖尿病人群中DR筛查的一致性，从而有助于通过早期检测和治疗减少并最终消除因DR导致的视力丧失。