Statistical Methods for Neural Mechanisms Mediating Cognitive System in Mental Health Research

心理健康研究中调节认知系统的神经机制的统计方法

• 批准号: 9145621
• 负责人: SEONJOO LEE
• 金额: $ 13.14万
• 依托单位: NEW YORK STATE PSYCHIATRIC INSTITUTE DBA RESEARCH FOUNDATION FOR MENTAL HYGIENE, INC
• 依托单位国家: 美国
• 财政年份: 2015
• 资助国家: 美国
• 起止时间: 2015-09-30 至 2019-05-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/9145621
• 关键词: Accounting; Adult; Affect; Age; Age-associated memory impairment; Aging; Alzheimer' s Disease; Applications Grants; Award; Biological Neural Networks; Biometry; Brain; Brain imaging; Cognition; Cognition Disorders; Cognitive; Cognitive aging; Data; Data Set; Data Sources; Development; Early Diagnosis; Elderly; Episodic memory; Functional Magnetic Resonance Imaging; Goals; Health; Heterogeneity; Image; Impaired cognition; Liquid substance; Machine Learning; Magnetic Resonance Imaging; Measurement; Measures; Mediating; Mediation; Mediator of activation protein; Mental Health; Mental disorders; Mentors; Methods; Modality; Modeling; Multicenter Studies; Nerve Degeneration; Neurosciences; Pathology; Pattern; Performance; Psychiatry; Psychopathology; Research; Research Training; Role; Speed; Statistical Methods; Structure; System; Techniques; Testing; Training; age effect; age related; base; brain volume; career; cerebral atrophy; cognitive ability; cognitive change; cognitive system; cognitive task; cognitive testing; design; method development; mild cognitive impairment; neuroimaging; neuromechanism; normal aging; novel; pathological aging; relating to nervous system; research and development; tool

 DESCRIPTION (provided by applicant): Psychiatric disorders, as well as normal aging, are often associated with a wide range of cognitive changes. Recent developments in neuroscience, particularly multimodal neuroimaging techniques, can provide better understanding of neural mechanisms that underlie these changes. Mediation analysis is a popularly used statistical method to investigate neural mechanisms. However, existing statistical methods were not designed to accommodate such large-scale, multi-dimensional, and complicated data in mediation analysis. The overarching aim of my K01 Mentored Research Development Award is to acquire training that will allow me to pursue a line of mental health research related to cognition and to develop novel statistical methods to support the emerging research in cognitive neuroimaging and mental health. I propose to receive training in: 1) cognition and its relationship with aging and psychopathology; 2) neural basis in cognition and multimodal neuroimaging; 3) functional mediation and selective mediation analysis. This will provide me with the tools to conduct research that will fill significant gaps in the statistical analyses on cognitive neuroimaging, particularly focusing on aging and neurodegeneration-associated cognitive decline: First, I propose to review and develop analytic tools to examine how aging and psychopathology affect cognitive system constructs. Understanding the patterns of age-related differences in each of these cognitive constructs and the possible heterogeneity related to the psychopathology is of great importance. I will test nonlinearity and heterogeneity in the latent cognitive constructs using two independent large data. Second, I will develop and examine activation networks (reference ability neural networks (RANNs)) associated with each of the cognitive constructs. Although it is critical to understand underlying neural networks of cognition, there are many unsolved problems in investigating the presence or structure of RANNs. I will develop a method to derive underlying networks, find RANNs taking into account possible spatial overlap among RANNs, and investigate the relationship between RANNs expression and performance on the concomitant cognitive constructs. Third, I will develop selective mediation analyses to examine the relationship between imaged neural substrates of aging and psychiatric disorder, such as regional volume loss, and cognitive decline and test if the different imaged brain modalities selectively mediate cognitive aging by influencing the expression of cognitive ability constructs. In totality, this training and research will inform hypotheses for an R01 grant application to be submitted in Year 4 of the award period. While to date I have exemplified a productive research career in biostatistics and psychiatry, I need substantially more training for successful completion of the research proposed in this application and to make a lasting contribution to the field of mental health.

 描述（由申请人提供）：精神疾病以及正常衰老通常与广泛的认知变化相关。神经科学的最新发展，特别是多模态神经成像技术，可以提供更好的理解这些变化背后的神经机制。中介分析是研究神经机制的一种常用统计方法。然而，现有的统计方法并不适合调解分析中如此大规模，多维和复杂的数据。我的K 01指导研究发展奖的总体目标是获得培训，使我能够从事与认知相关的心理健康研究，并开发新颖的统计方法，以支持认知神经成像和心理健康方面的新兴研究。我建议接受培训：1）认知及其与衰老和精神病理学的关系; 2）认知和多模态神经成像的神经基础; 3）功能调解和选择性调解分析。这将为我提供进行研究的工具，这些研究将填补认知神经成像统计分析中的重大空白，特别是关注衰老和神经退行性变相关的认知下降：首先，我建议审查和开发分析工具，以研究衰老和精神病理学如何影响认知系统结构。了解这些认知结构中与年龄相关的差异模式以及与精神病理学相关的可能异质性非常重要。我将使用两个独立的大数据来测试潜在认知结构中的非线性和异质性。其次，我将开发和研究与每个认知结构相关的激活网络（参考能力神经网络（RANN））。虽然理解认知的底层神经网络至关重要，但在研究RANN的存在或结构方面仍有许多未解决的问题。我将开发一种方法来推导底层网络，考虑到RANN之间可能的空间重叠，找到RANN，并研究RANN表达与伴随认知结构的性能之间的关系。第三，我将开发选择性调解分析，以检查成像的神经基板的老化和精神疾病，如区域体积损失，认知能力下降和测试之间的关系，如果不同的成像大脑模式选择性地介导认知老化的认知能力结构的影响表达。总的来说，这种培训和研究将告知R 01补助金申请的假设将在奖励期的第4年提交。虽然到目前为止，我已经在生物统计学和精神病学方面证明了一个富有成效的研究生涯，但我需要更多的培训才能成功完成本申请中提出的研究，并为心理健康领域做出持久的贡献。