Neuroimaging of Motor Imagery for Brain Computer Interface Applications

该奖项是根据2009年美国复苏和再投资法案（公法111-5）资助的。目前，美国有超过200万人患有不同程度的瘫痪，每年还有1.1万例新的脊髓损伤病例。因此，使这些人康复的手段将产生巨大的经济和社会影响。脑机接口的开发是为了“读取”这些人的思想，并将这些思想转化为通过计算机执行的行动，目的是通过为大脑提供新的输出途径来恢复瘫痪患者的功能。该项目的长期目标是开发一种新型的无创脑机接口系统，该系统可以可靠有效地执行复杂任务。这种系统的可用性将对帮助患有导致行动能力严重受损的神经系统疾病的患者产生重大影响。该项目的具体目标是使用功能性磁共振成像（MRI）和脑电图（即脑电波）记录，为与运动成像任务相关的个体特征提供坚实的知识基础，以显着改善基于运动成像的脑机接口系统的设计和实现。该项目的具体目标是：1)研究与各种运动意象任务相关的神经和血流动力学反应之间的空间共定位；2)开发适合成像“图像”大脑活动的超灵敏时空成像方法；3)基于从脑电波的空间、时间和频域提取的个体特征，开发和评估一种新的多维脑机接口系统。智力优势：这项拟议的研究解决了一个基本的科学挑战，即通过非侵入性测量将“思想”转化为行动。基于运动图像的脑机接口系统的一个关键组成部分是提取敏感信号，其中人类意图最好地编码。目前，此类签名主要基于头皮记录的脑电波在传感器空间的信息处理，需要受试者进行大量训练才能实现对多维环境的控制。头皮记录的脑电波缺乏特异性，与“意图”相关的神经元放电与记录的脑电波之间的关系仍然不明确。这项拟议研究的变革性在于将个人的“意图”直接映射到大脑皮层上，大脑皮层是神经信息处理的物理执行地。这种高度创新的方法将通过使用功能性核磁共振成像和脑波源成像在各种运动任务的想象中建立一对一的映射来实现。所提出的数据驱动的方法能够合理地融合功能性核磁共振信号和脑电波，从而产生对“思想”信号的复杂模式进行超灵敏成像的方法。该研究的成功完成将极大地扩展对运动图像神经机制的理解，并导致一种变革性的无创多维脑机接口系统的发展，该系统将显著提高特异性、效率和可靠性。更广泛的影响：拟议的项目有可能在几个领域产生重大的广泛影响：a)社会：目前在美国有超过200万人患有不同程度的瘫痪。这项提议的研究有望导致高度复杂的非侵入性脑机接口系统，这可能会极大地帮助这类患者的临床康复，有利于公共卫生和经济。b)技术性：拟议的研究解决了科学中的一个重大问题——解码与“思想”相关的信号以控制设备。本研究对神经科学、康复工程、控制理论、信号处理和成像科学等领域有重要贡献。c)加强研究和教育的基础设施：拟议的研究将有助于加强生物医学工程师、神经科学家和成像科学家之间的合作，并为研究生、博士后和本科生（包括少数民族和女学生）提供跨学科培训的独特机会。d)知识传播：研究结果将广泛传播给该领域的其他研究人员，包括学术出版物和软件代码的网站传播。

This award is funded under the American Recovery and Reinvestment Act of 2009 (Public Law 111-5)0933067HeThere are currently over two million people in the United States suffering from various degrees of paralysis with an additional eleven thousand new cases of spinal cord injury each year. A means to rehabilitate these individuals would thus have tremendous economic and social impact. The brain-computer interface has been developed as a means to 'read' the minds of these individuals and translate these thoughts into actions performed via a computer, which aims at restoring function in paralytics by providing the brain with new output pathways.The long-term goal of this project is to develop a novel non-invasive brain-computer interface system, which can perform complex tasks reliably and efficiently. The availability of such a system would have a significant impact in aiding patients with neurological disorders that cause significant impairment in mobility. The specific objective of the project is to use functional magnetic resonance imaging (MRI) and electroencephalogram (i.e. brain wave) recordings to provide a solid base of knowledge regarding the individual signatures associated with motor imagery tasks in order to significantly improve the design and implementation of motor imagery-based brain-computer interface systems. The specific aims of the project are to: 1) Investigate the spatial co-localizations between neural and hemodynamic responses associated with various motor imagery tasks; 2) Develop ultra-sensitive spatiotemporal imaging methods suited for imaging 'imagery' brain activity; 3) Develop and evaluate a novel multi-dimensional brain-computer interface system based on individual signatures extracted from the space, time and frequency domains of brain waves. Intellectual Merits: The proposed research tackles a fundamental scientific challenge in converting 'thoughts' into actions through noninvasive measurements. A key component of the motor imagery-based brain-computer interface systems is to extract sensitive signatures in which human intentions are best encoded. Currently, such signatures are mainly based on information processing of scalp-recorded brain waves at the sensor space, which require significant training by the subject to achieve control of a multi-dimensional environment. Scalp-recorded brain waves lack specificity and the relationship between neuronal firing associated with 'intentions' and the recorded brain waves remains ambiguous. The transformative nature of the proposed research is to map the 'intentions' of individuals directly over the cortex where neural information processing is physically being performed. This highly innovative approach will be realized by establishing one-on-one mapping during imagination of a variety of movement tasks through the use of functional MRI and brain wave source imaging. The proposed data-driven approach enables rational fusing of functional MRI signals and brain waves, leading to ultra-sensitive means imaging the complex patterns of 'thoughts' signals. The successful completion of the proposed research will greatly expand the understanding of the neural mechanisms of motor imagery, and lead to the development of a transformative noninvasive multi-dimensional brain-computer interface system, which would have significantly enhanced specificity, efficiency, and reliability. Broader Impacts: The proposed project has the potential for having major broad impacts in several areas: a) Societal: There are currently over two million people in the United States suffering from various degrees of paralysis. The proposed research promises to lead to highly complex noninvasive brain-computer interface systems, which may significantly aid in the clinical rehabilitation in this group of patients, benefiting public health and the economy. b) Technical: The proposed research addresses a significant problem in science - decoding signals related to 'thoughts' to control devices. The proposed research may make a significant contribution to neuroscience, rehabilitation engineering, control theory, signal processing, and imaging science. c) Enhancement of infrastructure for research and education: The proposed study will help strengthen collaborations among biomedical engineers, neuroscientists, and imaging scientists, and provide unique opportunities for interdisciplinary training of graduate students, postdoctoral fellows, and undergraduate students, including minority and female students. d) Knowledge dissemination: The findings will be broadly disseminated to other researchers in the field including scholarly publications and website dissemination of software codes.