Snapshot Retinal Imaging Mueller Matrix Polarimeter

描述（由申请人提供）：全球有超过6500万人患有青光眼。由于绝大多数患者在发病较晚时才会出现青光眼症状，因此早期诊断和治疗对于保持视力至关重要。目前，青光眼的诊断是基于视盘结构改变和功能性视野改变的结合。不幸的是，这两种测试都是相对主观的，因此青光眼的明确诊断可能会延迟或错过。然而，人们普遍认为，视网膜神经纤维层的缺陷（RNFL，即结构改变）比视野丧失的发病早6年。许多成像方法，包括共聚焦扫描激光检眼镜（cSLO），光学相干断层扫描（OCT）和扫描激光偏振法（SLP）已经开发出来，可以早期发现这些细微的结构缺陷。其中，SLP为实现青光眼引起的视网膜改变的早期检测提供了希望。然而，由于SLP不是一个完整的偏振计，由于RNFL的双衰减和退极化，其测量结果容易出现误差。因此，SLP的延迟分布的准确性受到未知的视网膜双衰减和去极化的影响，降低了设备的灵敏度和特异性。此外，商用slp只能成像视神经头附近的一小部分视网膜，受到运动伪影的影响，而且价格昂贵（使用商用现货（COTS）零件和机械扫描仪估计成本为1.7万美元）。我们已经克服了这些限制，使用更便宜的（使用COTS部件估计为1万美元），完整的穆勒矩阵（MM）偏光计，没有移动部件：快照视网膜成像穆勒矩阵（SRIMM）偏光计。该设备避免了时间扫描，并允许在大视野范围内实时测量视网膜的完全偏振状态。这个装置有几个优点。首先，该设备的成本降低可能使其成为筛查患者的更实用和更具成本效益的工具。其次，由于该设计没有时间扫描，因此可以在单个眼底快照中观察到大面积的视网膜区域，而无需复杂的图像配准算法。第三，能够获得具有高时间分辨率的完整MM数据集，通过消除与去极化和双衰减相关的测量误差，将提高该设备的临床敏感性和特异性。这是北卡罗莱纳州立大学、南加州大学眼科研究所和亚利桑那大学的合作成果。具体目标是：1。设计，校准，验证和测试眼底相机原型的快照视网膜成像穆勒矩阵（SRIMM）偏振计。2. 设计并建立一个模型眼，以表征和验证偏振计的性能。3. 通过高分辨率SD-OCT和蔡司GDx的对比测量，在兔模型和有限数量的人类患者中验证SRIMM偏振计的性能。

DESCRIPTION (provided by applicant): Worldwide, over 65 million people suffer visual impairment from glaucoma. Because the vast majority of patients rarely experience symptoms of glaucoma until relatively late in the disease, early diagnosis and treatment is essential in preserving vision. Currently, the diagnosis of glaucoma is based on a combination of structural optic disc changes and functional visual field changes. Unfortunately, both of these tests are relatively subjective and definitive diagnosis of glaucoma can be delayed or missed as a result. However, it is well accepted that defects in the retinal nerve fiber layer (RNFL; i.e. structural changes) precede the onset of visual field loss by as much as six years. A number of imaging methods including confocal scanning laser ophthalmoscopy (cSLO), optical coherence tomography (OCT) and scanning laser polarimetry (SLP) have been developed to allow early detection of these subtle structural defects. Of these, SLP provides promise in achieving the earliest detection of glaucoma-induced retinal changes. However, since SLP is not a complete polarimeter, its measurements are prone to error due to the RNFL's diattenuation and depolarization. Thus, accuracy of the SLP's retardance distribution is influenced by the unknown retinal diattenuation and depolarization, decreasing the device's sensitivity and specificity. Furthermore, commercial SLPs only image small sections of the retina in the neighborhood of the optic nerve head, suffer from motion artifacts, and are expensive (estimated cost of $17k using commercial off-the-shelf (COTS) parts and mechanical scanners). We have overcome these limitations with a less expensive (estimated $10k using COTS parts), complete Mueller matrix (MM) polarimeter with no moving parts: the Snapshot Retinal Imaging Mueller Matrix (SRIMM) polarimeter. This device avoids temporal scanning and allows a real-time measurement of the retina's complete polarization state over a large field of view. This device has several advantages. First, the device's cost reduction may make it a more practical and cost-effective instrument for screening patients. Second, since the design has no temporal scanning, large retinal areas can be observed, in a single fundus snapshot, without the need for complex image registration algorithms. Third, the ability to acquire a complete MM data set with high temporal resolution will improve the clinical sensitivity and specificity of this device by removing measurement errors related to depolarization and diattenuation. This is a collaborative effort between North Carolina State University, USC Eye Institute, and the University of Arizona. The specific aims are 1. Design, calibrate, validate and test a fundus camera prototype of a Snapshot Retinal Imaging Mueller Matrix (SRIMM) polarimeter. 2. Design and build a model eye to characterize and validate the performance of the polarimeter. 3. Validate the SRIMM polarimeter's performance in a rabbit model and in a limited number of human patients via comparative measurements taken with a high-resolution SD-OCT and a Zeiss GDx.