Li Hao, Liu Wenzhong, Dong Biqin, Kaluzny Joel V, Fawzi Amani A, Zhang Hao F
Department of Biomedical Engineering, Northwestern University, Evanston, IL 60208, USA.
Department of Ophthalmology, Northwestern University, Chicago, IL 60611, USA.
J Biophotonics. 2017 Jun;10(6-7):830-839. doi: 10.1002/jbio.201600053. Epub 2016 Jul 19.
Hyperspectral retinal imaging captures the light spectrum from each imaging pixel. It provides spectrally encoded retinal physiological and morphological information, which could potentially benefit diagnosis and therapeutic monitoring of retinal diseases. The key challenges in hyperspectral retinal imaging are how to achieve snapshot imaging to avoid motions between the images from multiple spectral bands, and how to design a compact snapshot imager suitable for clinical use. Here, we developed a compact, snapshot hyperspectral fundus camera for rodents using a novel spectral resolving detector array (SRDA), on which a thin-film Fabry-Perot cavity filter was monolithically fabricated on each imaging pixel. We achieved hyperspectral retinal imaging with 16 wavelength bands (460 to 630 nm) at 20 fps. We also demonstrated false-color vessel contrast enhancement and retinal oxygen saturation (sO ) measurement through spectral analysis. This work could potentially bring hyperspectral retinal imaging from bench to bedside.
高光谱视网膜成像可捕捉每个成像像素的光谱。它提供了光谱编码的视网膜生理和形态学信息,这可能有助于视网膜疾病的诊断和治疗监测。高光谱视网膜成像的关键挑战在于如何实现快照成像以避免多个光谱带图像之间的运动,以及如何设计适合临床使用的紧凑型快照成像仪。在此,我们使用新型光谱分辨探测器阵列(SRDA)开发了一种用于啮齿动物的紧凑型快照高光谱眼底相机,在该阵列的每个成像像素上单片制造了薄膜法布里-珀罗腔滤波器。我们以20帧/秒的速度实现了16个波长带(460至630纳米)的高光谱视网膜成像。我们还通过光谱分析展示了伪彩色血管对比度增强和视网膜氧饱和度(sO)测量。这项工作有可能将高光谱视网膜成像从实验室带到临床。