Three-dimensional structure-function relationships in glaucoma from SD-OCT

描述（由申请人提供）：本二期课题旨在阐明青光眼患者视野阈值敏感性与视网膜层、视神经头及连接神经纤维束结构特征之间的结构-功能关系。这一建议扩展了我们成功的第一阶段研究，建立了视网膜定量指标之间的结构-结构关系。在第二阶段，光谱域光学相干断层扫描（SD OCT）图像分析的进一步发展将产生新的损伤指标，有望更好地与视野阈值灵敏度相对应。该项目是由一个重要的临床问题驱动的——作为青光眼视网膜神经节细胞及其轴突不可逆损伤的测量，视野的可靠性和可重复性差。我们的长期愿景仍然是通过oct的客观结构测量来确定/预测青光眼的视觉功能。一旦实现，这种新方法将提供一种客观和可重复的测量方法，补充青光眼损伤的主观功能评估，减少因视觉反应长期波动而频繁进行视野测试的需要，并基于可靠的进展标记改善青光眼治疗。具体来说，我们将开发越来越复杂的预测模型，以产生青光眼损害的患者特异性预测模型。推动这项研究的最重要的假设是，利用神经纤维束轨迹对整个视网膜神经节细胞-轴突复合体形态进行新的定量度量，可以可靠地预测视觉功能。我们确定了以下具体目标：目标1：通过比较24-2阈值与青光眼和正常受试者的注册多场SD-OCT扫描得出的相应结构指数，为Humphrey 24-2视力测试（24度半径视野）所覆盖的视网膜的局灶结构-功能相关性建立基线。从视网膜内层（由视网膜神经节细胞和神经纤维层组成）的结构特性中推导出功能的基线预测模型。目的2：证明结合基于SD-OCT图谱的视网膜神经节细胞-轴突复合体（RGC-AC）轨迹的结构参数可以提高预测结构-功能模型的性能。目的3：评估基于个体的RGC-AC轨迹而非RGC-AC图谱是否能改善24-2阈值的预测。本研究将使用100名青光眼患者和40名年龄匹配的正常人在同一天使用7眼3D SD-OCT图像以及24-2视野测试数据进行。

DESCRIPTION (provided by applicant): This Phase-II proposal seeks to elucidate the structure-function relationships between visual field threshold sensitivity and the structural features of the retinal layers, the optic nerve head, and the connecting nerve fiber bundles in glaucoma. This proposal extends our successful Phase-I research that established structure- structure relationships among retinal quantitative indices. In this Phase II, further advancements of spectral domain optical coherence tomography (SD OCT) image analysis will yield new damage metrics expected to correspond much better with visual field threshold sensitivity. The project is driven by an important clinical problem - the poor reliability and reproducibility of th visual field as a measure of irreversible damage to the retinal ganglion cells and their axons in glaucoma. Our long-term vision remains to determine/predict glaucoma visual function from objective structure measurements by OCT. Once achieved, this new approach will provide an objective and reproducible measure, complementing subjective functional assessment of glaucoma damage, decrease the need for frequent visual field testing resulting from long-term fluctuation of visual response, and improve glaucoma treatment based on reliable progression markers. Specifically, we will develop predictive models of increasing complexity yielding a patient- specific predictive model of glaucomatous damage. The overriding hypothesis motivating the proposed research is that novel quantitative metrics of the entire retinal ganglion cell-axonal complex morphology utilizing the nerve fiber bundle trajectories allow reliable prediction of visual function. We have identified the following specific aims: Aim 1: Establish a baseline for the focal structural-functional correlation in the retina covered by the Humphrey 24-2 perimetry test (24 degree radius visual field) by comparing 24-2 thresholds with their corresponding structural indices derived from registered multi-field SD-OCT scans in glaucoma and normal subjects. Derive a baseline predictive model of function from structural properties of the inner retinal layers, comprised of retinal ganglion cell and nerve fiber layers. Aim 2: Demonstrate that incorporating structural parameters along SD-OCT atlas-based retinal ganglion cell- axonal complex (RGC-AC) trajectories improves the performance of the predictive structure-function model. Aim 3: Evaluate whether prediction of 24-2 thresholds is improved by deriving individual-based RGC-AC trajectories instead of from an RGC-AC atlas. The proposed work will be performed using 7-field per eye 3D SD-OCT images accompanied by 24-2 visual field test data on the same day from 100 patients with glaucoma and 40 age-matched normal subjects.