基于数字微镜的自适应光学活体视网膜神经节细胞成像系统及其在青光眼诊疗中的应用研究

Glaucoma is characterized by the irreversible damage of retinal ganglion cells (RGCs) and their axons. Early detection and early intervention are critical for glaucoma management. However, all current glaucoma diagnostic tools cannot directly reveal the early morphological changes of RGCs. In vivo imaging of RGCs could directly evaluate the earliest signs of glaucomatous morphological changes and the courses of cell damage during disease progression. Nevertheless, the transparency of RGCs poses a major challenge for imaging these cells in the living eyes. RGCs labeled by exogenous fluorophores have been imaged in animal eyes using the fluorescence adaptive optics scanning laser ophthalmoscopy (AOSLO) imaging system. But the toxicity of fluorescent markers and the invasive way in which they are administered make this method unsuitable for use in humans. The adaptive optics confocal phase imaging technology we previously developed has demonstrated the feasibility of improving the contrast of cells in images. Here we propose an innovative differential intensity contrast imaging technology based on digital micromirror device (DMD) using two split-detectors. The phase difference produced by light passing through the cells causes light intensity difference. DMD allows the detection of light intensity difference in all directions so that the contrast of the RGCs with transparent and uneven structures will be enhanced. Therefore, the DMD-AOSLO system will be able to directly image RGCs in vivo without the use of exogenous fluorophores. The application of this imaging system will facilitate the elucidation of the damage pattern of RGCs in monkey models of chronic ocular hypertension. An evaluation system based on the microstructures of human RGCs will be established to find new biomarkers for early diagnosis of glaucoma and biomarkers related to glaucoma progression. New evidences for the early diagnosis and early intervention of glaucoma are expected from our study.

青光眼的本质是视网膜神经节细胞（RGC）与轴索的不可逆损害，早期诊断和干预极为重要，但现有诊断手段均不能直接反映RGC的早期改变。RGC活体成像可直观评价其在疾病最早期和进展过程中的形态变化，但无色透明的RGC使其活体成像非常困难。自适应光学扫描激光检眼镜（AOSLO）可借助荧光标记物对动物RGC成像，但荧光物质的毒性和有创的给药方式令其不适用于人体。我们前期开发的自适应光学共焦相位成像技术展示了增强细胞图像对比度的能力。在此基础上，本课题拟研发基于数字微镜器件（DMD）的分探测器差分光强成像技术，构建DMD-AOSLO系统，在多个方向探测光经细胞界面时相位变化导致的光强差以实现不借助外源造影剂的活体RGC超微成像。应用该系统阐明慢性高眼压下猴眼RGC的损伤机制；建立基于人RGC超微结构的评价体系，探寻青光眼早期诊断和与疾病进展相关的新指标，以期对青光眼早期诊断、干预提供精准的科学依据。

青光眼的本质是视网膜神经节细胞（RGC）与轴索的不可逆损害，早期诊断和干预极为重要，但现有诊断手段均不能直接反映RGC的早期改变。RGC活体成像可直观评价其在疾病最早期和进展过程中的形态变化，从本质上加深对该病机制的理解，但无色透明的RGC使其活体成像非常困难。自适应光学辅助眼底成像是迄今唯一能实现活体视网膜单细胞水平分辨率成像的方法，目前较常见的应用领域为活体光感受器细胞成像。借助于玻璃体腔/视网膜下注射外源性的荧光标记物，自适应光学扫描激光检眼镜（AOSLO）可对动物RGC成像，但荧光物质的毒性和有创的给药方式令其不适用于人体。本项目首先进一步夯实了研究基础：在我们前期发现的节细胞-内丛状层 (GCIPL) 厚度对青光眼的总体诊断效能与视乳头周围视网膜神经纤维层(RNFL)厚度相比并无显著提升的基础上，基于光相干断层成像（OCT），首次发现了GCIPL在鉴别视野前期青光眼和非青光眼性高度近视视盘改变的效能优于其它所有OCT参数；首次发现了正常眼压性青光眼较原发性开角型青光眼的GCIPL局限性损害规律；分析了汉族人GCIPL厚度的影响因素并揭示了GCIPL厚度随年龄变化的独特规律。其次，本项目在青光眼及视网膜变性性疾病中发现了黄斑区神经节细胞层（GCL）容积、光感受器细胞层容积在病程进展中的变化规律，以及两者间的联动关系，证实了光感受器细胞的变化可反映部分RGC的变化。另外，本项目改进了原有AOSLO系统的光电结构，使其成像宽容度得到改善、成像质量得到提升、图像后期处理方式得到改进；基于上述发现，通过观测光感受器细胞的超微结构反应RGC的状态。更进一步，基于共焦相位、差分光强技术，在进行RGC观测的过程中，获取了内层视网膜毛细血管中活体单个红细胞的清晰影像，在此基础上分析青光眼及视网膜变性性疾病的视网膜毛细血管血流改变。最后，将基于自适应光学超微成像的结果与较易开展临床应用的RGC间接评价指标进行相关性分析与量化评估，肯定了最小GCIPL厚度和GCL容积对青光眼早期诊断的价值。