System for Increasing Patient Access to Eye Exams for Diabetic Retinopathy

描述（由申请人提供）：拟议SBIR计划的总体目标是在糖尿病视网膜病变（DR）筛查系统中实施和评估低成本视网膜摄像机，即激光扫描数码摄像机（LSDC）。未经治疗的DR已在全球服务不足的人群中达到流行水平。由于LSDC已在初步研究中证明其准备好进行DR筛查，因此提出了快速通道建议。用于视网膜成像的现有仪器已经不足以用于DR筛查，因为它们太昂贵或者提供太高比例的不可分级图像。与这些设备不同的是，专利LSDC设计提供了一个非散瞳设计和高对比度视网膜图像的组合，但仍然是低成本和易于操作。舒适的近红外照明提供了从视网膜的强光返回，而不管眼睛的颜色。前节和后节成像是实时的。这些功能使LSDC能够筛查小瞳孔、深色眼睛或中度白内障患者，同时提供足够宽的视网膜视野进行分析。LSDC将由永旺成像公司进一步开发，使用加州大学伯克利分校和印第安纳州大学提供的现场测试数据和分析。已经很低的成本，LSDC材料和组装成本的节省将进一步探索。除了较低的仪器成本外，采用经过培训的设备操作员和分级员的支持系统，包括具有成本效益和准确的图像传输协议，对于DR筛查的整体成功至关重要。因此，永旺将与加州大学伯克利分校运营的现有临床通信系统EyePACS合作，该系统已经为服务不足的个人提供DR筛查和远程医疗转诊服务。在第一阶段，永旺与印第安纳州大学、加州大学伯克利分校和Vision for All合作，将使用经过培训但非专业的操作员调查LSDC的性能。初始筛选将基于特征，即，将分析使用LSDC检测特定DR特征的成功率。LSDC将与商用非散瞳眼底相机和散瞳眼底检查进行比较。在第二阶段，永旺将进一步降低成本，并将第一阶段研究的操作员/患者反馈纳入第二代LSDC设计。永旺还将培训非专业的图像分级人员，以识别LSDC图像中发现的DR特征，并确定他们是否可以准确地指定哪些患者需要转诊进行后续眼科护理。一项比较研究将确定使用经过培训的操作员进行LSDC筛查、使用EyePACS进行远程医疗图像传输以及经过培训的图像分级员是否与使用昂贵设备和专家图像分析做出的独立转诊决定相匹配。在这项研究的结论，永旺将有现场测试的LSDC和低成本的筛选系统，建立其能力，相对于昂贵的方法。如果成功，永旺将准备扩大DR筛查业务，并帮助减少健康差距。

DESCRIPTION (provided by applicant): The overall goal of the proposed SBIR program is to implement and evaluate a low cost retinal camera, the Laser Scanning Digital Camera (LSDC), in a system for diabetic retinopathy (DR) screening. Untreated DR has reached epidemic levels in underserved populations worldwide. As the LSDC has already demonstrated its readiness for DR screening in preliminary studies, a Fast-Track proposal is presented. Existing instruments for retinal imaging have been inadequate tools for DR screening as they are too expensive or provide too high a proportion of ungradeable images. Unlike these devices, the patented LSDC design provides a combination of a non-mydriatic design and high contrast retinal images, yet remains low cost and easy to operate. Comfortable near infrared illumination provides strong light return from the retina regardless of eye color. Anterior and posterior segment imaging are real-time. These features allow the LSDC to screen patients with small pupils, darkly pigmented eyes, or moderate cataracts, while providing a sufficiently wide retinal field for analysis. The LSDC will be further developed by Aeon Imaging, using field-testing data and analysis provided by UC Berkeley and Indiana University. Already low cost, LSDC materials and assembly cost savings will be explored further. Apart from low instrument costs, a supporting system that employs trained device operators and graders, including cost-effective and accurate image transfer protocols, is critical for overall DR screening success. Thus, Aeon will partner with an existing clinical communications system, EyePACS, operated by UC Berkeley, that is already providing DR screening and telemedicine referral services to underserved individuals. In Phase I, Aeon, in collaboration with Indiana University, UC Berkeley, and Vision for All, will investigate the performance of the LSDC using trained but non-specialist operators. The initial screening will be feature- based, i.e. the success in detecting specific DR features with the LSDC will be analyzed. The LSDC will be compared against a commercial non-mydriatic fundus camera and a dilated fundus exam. In Phase II, Aeon will incorporate further cost reductions and operator/patient feedback from Phase I studies into the 2nd generation LSDC design. Aeon will also train non-specialist image graders to identify DR features found in LSDC images, and determine whether they can accurately specify which patients require referral for follow-on eye care. A comparative study will determine if screening with the LSDC using trained operators, telemedicine image transfer with EyePACS, and trained image graders matches the independent referral decision made using costlier devices and specialist image analysis. At the conclusion of this study, Aeon will have field-tested the LSDC and the low cost screening system, establishing its capability with respect to costlier approaches. If successful, Aeon will be prepared to expand DR screening operations and help reduce health disparities.