Training in lesion-symptom mapping for speech-language research

用于言语研究的病变症状映射培训

• 批准号: 9040405
• 负责人: Julie A Fiez
• 金额: $ 16.79万
• 依托单位: UNIVERSITY OF PITTSBURGH AT PITTSBURGH
• 依托单位国家: 美国
• 财政年份: 2016
• 资助国家: 美国
• 起止时间: 2016-05-18 至 2018-04-30
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/9040405
• 关键词: Adult; Aphasia; Big Data; Brain; Brain imaging; Broca Aphasia; Chronic; Collaborations; Communication impairment; Communities; Complex; Computer software; Data; Data Science; Data Set; Educational workshop; Evaluation; Gold; Health; Image; Interdisciplinary Study; K-Series Research Career Programs; Knowledge; Language; Language Disorders; Leadership; Learning; Lesion; Machine Learning; Manuals; Maps; Mentors; Methods; Neuropsychology; Participant; Pennsylvania; Population; Registries; Reproducibility; Research; Research Activity; Research Methodology; Research Personnel; Resources; Sample Size; Seeds; Slice; Speech; Stroke; Structure-Activity Relationship; Symptoms; Syndrome; Techniques; Testing; Tissues; Training; Travel; United States; United States National Institutes of Health; Videoconferences; Videoconferencing; Visit; Work; abstracting; base; career; data sharing; design; disability; experience; improved; interest; knowledge base; language processing; method development; named group; neuropsychological; novel; patient registry; programs; relating to nervous system; research and development; research study; skills; speech processing; stroke survivor; success; theories

 DESCRIPTION (provided by applicant): Training in lesion-symptom mapping for speech-language research Abstract: Researchers rely upon the lesion method to evaluate the speech-language status of stroke survivors and draw inferences about underlying brain function. This use of neuropsychology is highly valued in basic speech- language research because it can support causal inferences about brain structure/function relationships. Crucially, advances in analytic techniques and brain image computing are creating a new landscape for neuropsychological research. In this new landscape, the lesion method represents a form of big-data science that requires large sample sizes and complex image computing to implement lesion-symptom mapping (LSM) across the entire brain, without prior regions of interest. Expertise in these new techniques is becoming critical for high impact speech-language research. The career enhancement plan will provide the candidate with training in cutting-edge LSM. The candidate is an established speech-language investigator with a basic program of multidisciplinary research that includes populations with communication disorders due to stroke. The career enhancement will come at an ideal point, because it will build on the candidate's success in establishing an open-access research registry of stroke survivors (the Western Pennsylvania Patient Registry, WPPR), and current work to develop and validate collaborative videoconferencing for remote neuropsychological assessment. These efforts have created the recruitment pool and datasets that are needed for LSM. The career enhancement will provide the training needed to leverage these resources, thereby augmenting the candidate's program of research and career trajectory. The overarching objectives are to: (1) retool the skills of the candidate to infuse LSM into her program of speech-language research, (2) seed data sharing and data science partnerships to boost the candidate's leadership of WPPR as a national resource, and (3) advance understanding of LSM methods and the neural substrates for speech and language to improve the knowledge base of the candidate and other investigators. The candidate proposes a synergistic set of activities. Didactic activities will give training in machie learning and brain image computing, scholarly travel experiences will afford opportunities to network with speech-language researchers and data scientists whose work is relevant for LSM, and two research studies will provide a hands-on opportunity for the candidate to acquire, apply, and extend LSM methods under the guidance of a superb mentoring team. Study 1 will use univariate and multivariate LSM analysis to investigate the neural substrates of chronic Broca's aphasia and the factors that influence the reproducibility of LSM results. Study 2 will develop and evaluate a workflow for automated lesion segmentation, using a software platform (3D Slicer) that involves two NIH-supported data science centers. Overall, the career enhancement will retool the skills, research network, and knowledge base of an established investigator, allowing the candidate to significantly augment her program of speech-language research and advance the utility of WPPR as a national resource for speech-language research.

 描述（由申请人提供）：培训的病变症状映射的语言研究摘要：研究人员依靠病变的方法来评估中风幸存者的语言状态，并推断潜在的脑功能。神经心理学的这种应用在基础言语语言研究中受到高度重视，因为它可以支持关于大脑结构/功能关系的因果推论。至关重要的是，分析技术和大脑图像计算的进步正在为神经心理学研究创造一个新的景观。在这种新的情况下，病变方法代表了一种大数据科学，需要大样本量和复杂的图像计算来实现整个大脑的病变-症状映射（LSM），而无需先前的感兴趣区域。这些新技术的专业知识对于高影响力的语音语言研究变得至关重要。职业提升计划将为候选人提供最先进的LSM培训。候选人是一名成熟的语音语言研究者，拥有多学科研究的基本计划，包括因中风而导致沟通障碍的人群。职业提升将在一个理想的点来，因为它将建立在候选人的成功建立一个开放访问的研究登记中风幸存者（西宾夕法尼亚州患者登记处，WPPR），以及目前的工作，以开发和验证远程神经心理评估的协作视频会议。这些努力建立了LSM所需的征聘人才库和数据集。职业提升将提供利用这些资源所需的培训，从而增强候选人的研究和职业轨迹计划。总体目标是：（1）重新调整候选人的技能，将LSM注入她的语音语言研究计划，（2）种子数据共享和数据科学合作伙伴关系，以提高候选人对WPPR作为国家资源的领导力，以及（3）推进对LSM方法和语音和语言神经基质的理解，以提高候选人和其他研究人员的知识基础。候选人提出了一套协同增效的活动。教学活动将提供Machie学习和大脑图像计算方面的培训，学术旅行经验将提供与语音语言研究人员和数据科学家建立联系的机会，他们的工作与LSM相关，两项研究将为候选人提供动手的机会，以便在一流的指导团队的指导下获得，应用和扩展LSM方法。研究1将使用单变量和多变量LSM分析来研究慢性Broca失语症的神经基质以及影响LSM结果重复性的因素。研究2将使用涉及两个NIH支持的数据科学中心的软件平台（3D Slicer）开发和评估自动病变分割的工作流程。总体而言，职业提升将重组技能，研究网络和既定调查员的知识基础，使候选人能够显着增强她的语音语言研究计划，并推进WPPR作为语音语言研究国家资源的实用性。