Novel ultrahigh speed swept source OCT angiography methods in diabetic retinopathy

糖尿病视网膜病变的新型超高速扫源 OCT 血管造影方法

基本信息

批准号：
10656644
负责人：
JAMES G FUJIMOTO
金额：
$ 50.07万
依托单位：
MASSACHUSETTS INSTITUTE OF TECHNOLOGY
依托单位国家：
美国
项目类别：
财政年份：
2023
资助国家：
美国
起止时间：
2023-08-01 至 2028-06-30
项目状态：
未结题

来源：
https://reporter.nih.gov/project-details/10656644
关键词：
3-Dimensional Acceleration Address Age Aging Angiography Area Artificial Intelligence Assessment tool Background Diabetic Retinopathy Biological Markers Blindness Blood Vessels Blood capillaries Blood flow Clinical Clinical Management Clinical Trials Complications of Diabetes Mellitus Computer software Cross-Sectional Studies Data Data Set Detection Diabetic Retinopathy Disease Disease Progression Early Diagnosis Extravasation Eye Generations Goals Image Individual Intervention Ischemia Lasers Location Measures Mediating Medical Methods Microaneurysm Monitor Motion Noise Optical Coherence Tomography Outcome Papillary Pathogenesis Patients Performance Peripheral Prediction of Response to Therapy Protocols documentation Research Resolution Retina Retinal Diseases Risk Role Sampling Scanning Severities Severity of illness Signal Transduction Source Speed Structure Surface Surrogate Markers Techniques Technology Thrombosis Time Vascular Endothelial Growth Factors Vascular Permeabilities Vision Visit Visualization age group aging population biomarker identification cohort deep learning diabetes management diabetic diabetic patient disease mechanisms study drug development glycemic control high risk image processing imaging biomarker imaging modality improved inhibitor therapy innovation macula macular edema multidisciplinary neovascular neovascularization next generation novel predicting response predictive marker preservation prevent programs proliferative diabetic retinopathy response risk prediction tool treatment response vascular abnormality vessel regression

项目摘要

Diabetic retinopathy (DR) is a leading cause of blindness in working age populations. Glycemic control remains the best strategy for preventing or slowing disease progression. However, early detection of vision threatening retinal changes and prompt treatment is necessary for preserving visual function in more advanced stages. Clinical management of DR also requires objective tools for assessing retinal structure and microvascular abnormalities, as well as methods to evaluate treatment response. The goal of this multidisciplinary program is to develop novel optical coherence tomography (OCT) imaging methods and biomarkers that can identify and characterize retinal microvasculature and blood flow alterations in DR, which can be used for early disease detection, predicting progression, monitoring treatment response, and improving clinical trial efficiency. We will develop next generation OCT angiography (OCTA) technology and analysis frameworks (Aim 1). This hardware- software innovation will develop ultrahigh speed swept-source OCT (SS-OCT) to enable increased dynamic range OCTA characterization of retinal blood flow, higher definition volumetric data to visualize vasculature at the capillary level, and enable wide field structural and vascular imaging. These advances will develop a quantitative surrogate marker for capillary level blood flow speeds in retinal microvasculature. The ability to assess blood flow speeds at the capillary level represents a paradigm shift in OCTA imaging, enabling detection of subtle vascular flow alterations that occur at earlier disease stages, which are related to pathogenesis of diabetic macular edema (DME), capillary non-perfusion and neovascularization. New software motion correction technology can correct volumetric OCT/OCTA data for eye motion in three dimensions and artificial intelligence / deep learning techniques can generate reliable layer segmentations for OCTA, to facilitate tracking longitudinal changes across multiple patient visits. Using ultrahigh speed SS-OCT and OCTA blood flow biomarkers, we will perform a cross-sectional study in a cohort of diabetic patients and age-matched healthy controls (Aim 2). The study will evaluate repeatability of OCTA-driven VISTA flow markers, investigate if OCTA blood flow biomarkers are correlated with DR severity, and if they are associated with DME, capillary non-perfusion or neovascularization. Finally, we will monitor DR eyes longitudinally to investigate the role of subtle blood flow alterations in disease progression and predicting treatment response (Aim 3). The hypothesis is that OCTA blood flow biomarkers can identify regions with abnormal blood flow, which are at increased risk of thrombosis and leakage, and hence DME, as well as regions that are ischemic, and at increased risk of neovascularization. Similarly, we will also investigate if OCTA blood flow biomarkers can be used to predict response to anti-VEGF (vascular endothelial growth factor) therapy, including DME resolution, neovascular regression, and/or treatment durability. If successful, this program has the potential to improve clinical DR management and help DR patients to achieve better vision outcomes.

糖尿病视网膜病变（DR）是工作年龄人群失明的主要原因。血糖控制仍然有效预防或减缓疾病进展的最佳策略。然而，及早发现威胁视力的视网膜变化和及时治疗对于在晚期阶段保留视功能是必要的。 DR 的临床管理还需要客观的工具来评估视网膜结构和微血管异常情况以及评估治疗反应的方法。这个多学科计划的目标是开发新型光学相干断层扫描（OCT）成像方法和生物标志物，可以识别和描述 DR 中视网膜微血管和血流变化的特征，可用于早期疾病检测、预测进展、监测治疗反应并提高临床试验效率。我们将开发下一代 OCT 血管造影 (OCTA) 技术和分析框架（目标 1）。这个硬件—— 软件创新将开发超高速扫频OCT (SS-OCT)，以实现更高的动态视网膜血流的 OCTA 表征范围，更高分辨率的体积数据，可将脉管系统可视化毛细血管水平，并实现广域结构和血管成像。这些进步将发展出视网膜微血管系统中毛细血管水平血流速度的定量替代标记。有能力评估毛细血管水平的血流速度代表了 OCTA 成像的范式转变，从而实现了检测疾病早期阶段发生的细微血管血流改变，这与疾病的发病机制有关糖尿病性黄斑水肿（DME）、毛细血管无灌注和新生血管形成。新软件运动校正技术可以纠正三维眼动的体积 OCT/OCTA 数据和人工智能 / 深度学习技术可以为 OCTA 生成可靠的层分割，以利于纵向跟踪多次就诊时发生的变化。使用超高速 SS-OCT 和 OCTA 血流生物标志物，我们将在一组糖尿病患者和年龄匹配的健康对照中进行横断面研究（目标 2）。这研究将评估 OCTA 驱动的 VISTA 血流标记物的重复性，调查 OCTA 血流生物标记物是否与 DR 严重程度相关，如果它们与 DME、毛细血管无灌注或新血管形成。最后，我们将纵向监测DR眼睛，以研究细微血流的作用疾病进展的改变和预测治疗反应（目标 3）。假设 OCTA 血液血流生物标志物可以识别血流异常的区域，这些区域血栓形成的风险增加，渗漏，因此 DME，以及缺血区域，并且新生血管形成的风险增加。同样，我们还将研究 OCTA 血流生物标志物是否可用于预测抗 VEGF 的反应（血管内皮生长因子）治疗，包括 DME 解决、新生血管消退和/或治疗耐用性。如果成功，该计划有可能改善临床 DR 管理并帮助 DR 患者以实现更好的视力结果。