SCH: INT: Collaborative Research: Detection, Assessment and Rehabilitation of Stroke-Induced Visual Neglect Using Augmented Reality (AR) and Electroencephalography (EEG)

单侧空间忽视是一种知觉障碍，是中风后右侧脑损伤最常见的后果之一，在全球1500万中风患者中，29%的人会发生这种疾病。疏忽的患者表现出对大脑受损部分对面的物体或事件的疏忽。他们经常错过盘子一侧的食物，错过页面一侧的单词，撞到左边的门框，被移动的物体搞糊涂，害怕走在拥挤的地方。目前检测和修复忽视的黄金标准缺乏对日常生活活动(ADL)中遇到的动态任务和背景的概括性。该项目的研究人员将开发一个脑机接口(BCI)系统，该系统将在增强现实(AR)环境中实现，用于检测、评估和康复ADL期间的单边忽视。更具体地说，该系统将实时监测通过脑电(EEG)记录的大脑活动，以检测和评估视觉上被忽视的非个人空间。此外，该系统还将包括触觉、听觉和视觉刺激，同时使用者在康复期间进行真实世界的任务，以减少与忽视相关的残疾。预计该项目的新科学发现和工程改进将对BCI的当前实践产生影响：(I)使这种系统能够在更自然的环境中设计和实施，提供更身临其境的体验；以及(Ii)扩大BCI在设计其他神经疾病的干预和康复技术方面的使用。本项目将促进STEM教育，为从K-12到研究生水平的研究人员提供严格的培训和各种实践经验。本项目的研究目标是介绍一个卒中疏忽检测、评估和康复系统的原型，其特点是：(1)在基于视觉诱发脑电的BCI的设计中无缝集成脑电和AR，以便在日常生活活动中操作；(2)通过贝叶斯推理模型进行准确和连续的脑电事件相关电位检测，以进行疏忽评估和地图绘制；(3)以日常生活活动为重点的疏忽干预的信息论优化设计；以及(Iv)多模式实时反馈，用于干预期间忽视相关残疾的康复。与常见的电脑疏忽评估方法不同，EEG不需要患者做出任何身体反应。此外，EEG的使用允许自动化，使其成为指导个性化和自动化忽视干预的理想方法。众所周知，在现有干预措施中，显示出减少忽视的希望的一个共同因素是对身体或环境被忽视的一侧的多模式刺激。在连续的脑电监测过程中，当检测到疏忽时，及时反馈给用户，将使这种刺激成为可能。最后，在急性住院康复期间，结合AR耳机和基于技能的训练使用，计划中的系统将提供在有意义的日常活动中提供高强度重复刺激和进步的机会。该项目的成果将通过技术报告、期刊出版物和会议报告向科学界传播。通过该项目开发的所有软件将通过档案库公开提供。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

Unilateral spatial neglect is a perceptual disorder that is one of the most common consequences of right-side brain damage after stroke, occurring in 29% of the 15 million people who sustain stroke worldwide. Patients with neglect demonstrate inattention to objects or events on the side that is opposite to the damaged part of the brain. They often miss food on one side of the plate, missing words on one side of the page, bumping into the left door jamb, getting confused by moving objects, and being fearful of walking in crowded places. The current gold standard for detecting and rehabilitating neglect lacks generalizability to dynamic tasks and contexts encountered during activities of daily living (ADL). The investigators in this project will develop a brain-computer interface (BCI) system that will be implemented in augmented reality (AR) environment for detection, assessment and rehabilitation of unilateral neglect during ADL. More specifically, the system will in real-time monitor the brain activity recorded through electroencephalography (EEG) for the detection and assessment of visually neglected extra-personal space. Moreover, the system will also include haptic, auditory and visual stimulation while the users are engaged in real-world tasks conducted during rehabilitation for reducing neglect-related disabilities. It is also anticipated that the novel scientific discoveries and engineering enhancements of this project will have effects on the current practice on BCIs: (i) enabling design and implementation of such systems in more naturalistic environments providing more immersive experiences; and (ii) expansion of the use of BCIs in the design of intervention and rehabilitation techniques for other neurological disorders. This project will promote STEM education and provide rigorous training and variety of hands-on experiences to researchers from K-12 to graduate level.The research objective of this project is to introduce a prototype for stroke-induced neglect detection, assessment, and rehabilitation system, featuring: (i) seamless integration of EEG and AR in the design of visually evoked EEG-based BCIs to operate during activities of daily living; (ii) accurate and continuous EEG event related potential detection for neglect assessment and mapping through Bayesian inference models; (iii) information theoretic optimum design of neglect intervention focusing on activities of daily living; and (iv) multimodal real-time feedback for rehabilitation of neglect related disabilities during intervention. Unlike the common computerized neglect assessment methods, EEG will not require any physical responses from the patient. Also, the use of EEG permits automation, making it an ideal method to guide a personalized and automated neglect intervention. It is known that one common element among the existing interventions that have shown promise for reducing neglect is multimodal stimulation to the neglected side of the body or environment. Timely feedback to the user when neglect is detected during the continuous EEG monitoring will enable this stimulation. Finally, used in conjunction with AR headset and skill-based training during acute inpatient rehabilitation, the planned system will provide the opportunity to deliver high-intensity repetitive stimulation with progression during meaningful everyday activities. The outcomes of this project will be disseminated to the scientific community through technical reports, journal publications and conference presentations. All software developed through this project will be publicly available through archival repositories.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.