Hyperspectral Imaging of the Normal and Age-related Macular Degeneration Fundus

描述（由申请人提供）：一个在眼科疾病、影像学、视网膜病理学和生物医学工程方面具有世界一流专业知识的团队将首次利用高光谱成像技术，通过组织病理学相关性验证，对年龄相关性黄斑变性（AMD）进行体内分子探测。他们的系统是一种特殊改良的眼底相机，利用尖端的生物医学图像分析技术，通过识别与AMD病变相关的各种发色团和荧光团的分布和光谱特征，将彻底改变该领域。Drusen是AMD的标志性病变，具有生物化学异质性，是了解这种疾病的关键。然而，即使是目前分辨率最高的技术，也无法在体内确定它们的组成。高光谱成像数据，现在独特的临床可实现与新的快照高光谱设备，提供了一个视觉的光谱活检的视网膜。这些数据被编码成一个四维“超立方体”，具有两个空间（x和y坐标）和两个光谱（振幅和波长）维度。这个复杂的数据立方体可以用先进的无监督数学工具来探索，这些工具可以在数据矩阵中搜索出主要的光谱特征。通过分析drusen和其他AMD病变的光谱反射率和自身荧光（AF）特征，本研究将实现这些病变的体内光谱分类。为了实现这一目标，研究小组计划获取所有表型AMD的高光谱、摄影、AF和红外扫描激光检眼镜（SLO）和光谱域光学相干断层扫描（SD-OCT）图像，并通过优化的矩阵分解协议对高光谱图像进行分析，以恢复优势光谱及其空间分布。高光谱自动对焦图像也将同时获得与同一设备通过添加适当的激发和屏障过滤器。此外，基因型供体眼睛的高光谱成像将提供与drusen光谱特征相关的drusen和其他AMD病变成分的组织病理学证实，以及强大的基因型/表型相关性。长期目标：我们设想一个集成的成像系统，AMD和眼科护理，结合这种范式转变的技术。拥有这样一个体内分子探针系统将有助于研究。由此获得的见解将对临床诊断和治疗具有很高的价值。此外，这样的系统将加速转化研究与前瞻性治疗药物的敏感和早期结果测试，拯救视力，从而为社会提供巨大的利益。

DESCRIPTION (provided by applicant): A team with world-class expertise in ophthalmic disease, imaging, retinal pathology, and biomedical engineering will utilize hyperspectral imaging to provide, for the first time, in vivo molecular probing, validated with histopathologic correlation, of age-related macular degeneration (AMD). Their system, a specially modified fundus camera harnessed to cutting-edge biomedical image analysis techniques, will revolutionize the field by identifying the distribution and spectral signature of the various chromophores and fluorophores associated with AMD lesions. Drusen, the hallmark lesions of AMD, are biochemically heterogeneous and are key to understanding this disease. However, their composition cannot be determined in vivo with even the current highest resolution techniques. Hyperspectral imaging data, now uniquely clinically attainable with a new snapshot hyperspectral device, offers a vision of spectral biopsy of the retina. This data is encoded in a four-dimensional "hypercube," with two spatial (x and y coordinates) and two spectral (amplitude and wavelength) dimensions. This complex data cube can be explored with advanced unsupervised mathematical tools that search out the dominant spectral signatures in the data matrix. By dissecting the spectral reflectance and autofluorescence (AF) signatures from drusen and other AMD lesions, this research will achieve in vivo spectral classifications of these lesions. To accomplish this, the research team plans to acquire hyperspectral, photographic, AF and infrared scanning laser ophthalmoscope (SLO), and spectral domain optical coherence tomography (SD-OCT) images of all phenotypes of AMD, with analysis of the hyperspectral images by an optimized matrix factorization protocol to recover dominant spectra and their spatial distributions. Hyperspectral AF images will also be acquired simultaneously with the same device by adding appropriate excitation and barrier filters. Further, hyperspectral imaging of genotyped donor eyes will provide histopathologic confirmation of the components of drusen and other AMD lesions that correlate with the drusen spectral signatures, as well as robust genotype/phenotype correlations. Long-term goals: We envision an integrated imaging system for AMD and ophthalmologic care that incorporates this paradigm shift in technology. Having such a system of in vivo molecular probes will be instrumental in research. The insights so obtained will be of high value in clinical diagnosis and treatment. In addition, such a system will accelerate translational research with sensitive and early outcome testing of prospective therapeutic agents, saving sight and thereby providing enormous benefit to society.