Low Latency Eye-Motion Compensation

低延迟眼动补偿

• 批准号: 10610887
• 负责人: Alfredo Dubra
• 金额: $ 43.64万
• 依托单位: STANFORD UNIVERSITY
• 依托单位国家: 美国
• 财政年份: 2021
• 资助国家: 美国
• 起止时间: 2021-06-01 至 2025-04-30
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/10610887
• 关键词: Address; Algorithms; Bypass; Clinical; Communities; Compensation; Computer software; Crystalline Lens; Data; Dedications; Diagnosis; Disease; Disease Progression; Early Diagnosis; Electronics; Eye; Eye Movements; Eye diseases; Failure; Friction; Functional Imaging; Glaucoma; Goals; Human; Image; Individual; Inner Nuclear Layer; Light; Location; Measurement; Measures; Mechanics; Microscopic; Mission; Modeling; Motion; Movement; Ocular Motility Disorders; Ophthalmoscopes; Ophthalmoscopy; Optics; Pathologic Nystagmus; Pathology; Play; Psychophysics; Pupil; Research; Residual state; Resolution; Retina; Rotation; Sampling; Scanning; Speed; Stimulus; Structure; System; Technology; Testing; Time; Visualization; adaptive optics; adaptive optics scanning laser ophthalmoscopy; cellular imaging; cost; human imaging; human subject; image registration; imaging modality; improved; instrument; instrumentation; meter; millisecond; optic nerve disorder; prediction algorithm; retinal imaging; software development; structural imaging

PROJECT SUMMARY Adaptive optics (AO) scanning ophthalmoscopy allows non-invasive visualization of microscopic retinal structures by correcting the optical blur that is unique to each eye. In these instruments, however, image distortion due to involuntary fixational eye movement is a barrier to improving the understanding, early diagnosing, and management of eye disease. We propose to develop high-bandwidth eye motion stabilization technology and image registration software to mitigate this distortion even in subjects with the most extreme forms of involuntary eye motion, including nystagmus. This will be achieved by using a high frame rate pupil tracker that measures the eye’s orientation and by steering galvanometric optical scanners to simultaneously stabilize the view of the pupil and the retina as seen by AO scanning ophthalmoscopes with sub-millisecond total latency. Our goal here is to develop low-cost, low-complexity, and high-bandwidth retinal imaging stabilization instrumentation that can be used to upgrade both AO and non-AO ophthalmoscopes, using control algorithms that incorporate eye motion prediction, electronic and mechanical latency. Finally, we will also develop image registration software for scanning ophthalmoscopes that uses the high frame rate pupil tracking to improve accuracy and reduce failure rate, as well as automate the identification of and un-distort reference frames. The proposed technology will be applicable to all scanning ophthalmoscopes, irrespective of imaging modality.

项目摘要 自适应光学器件（AO）扫描眼镜检查允许微观视网膜可视化 通过纠正每只眼睛独有的光学模糊来结构。但是，在这些乐器中，图像 由于非自愿的固定眼运动而引起的失真是提早改善理解的障碍 诊断和治疗眼病。我们建议开发高带宽眼运动稳定 技术和图像注册软件即使在最极端的受试者中也可以减轻这种失真 非自愿眼运动的形式，包括眼球震颤。这将通过使用高帧速率学生来实现 跟踪器可以测量眼睛的方向以及通过转向电量光学扫描仪简单 稳定AO扫描眼镜的AO扫描眼镜的视野和视网膜的视野 潜伏期。我们的目标是发展低成本，低复杂性和高带宽视网膜成像稳定 可以使用控制算法来升级AO和非AO眼镜的仪器 结合了眼运动预测，电子和机械潜伏期。最后，我们还将开发图像 使用高帧速率瞳孔跟踪来改进的扫描眼镜扫描的注册软件 准确性并降低故障率，以及自动化的识别和UN-DISTORT参考框架。 提出的技术将适用于所有扫描眼镜镜，而与成像方式无关。