Collaborative Research: Spatiotemporal Fractional Modeling of Blood-Oxygen-Level Dependent Signals

在大脑与世界相互作用的背景下了解大脑，是评估其在健康和疾病状态下的动态的关键。功能磁共振成像(FMRI)技术能够跟踪一段时间内的血氧水平依赖信号，并作为神经活动的替代。目前的方法有局限性，因为它们假设不同大脑区域之间的相互依赖在整个记录期是恒定的。这些方法还假设大脑是一个孤立的系统，不考虑外部刺激。为了捕捉高度复杂的动力学并使人们能够理解人类认知的神经基础，本研究项目将开发新的数据分析方法来处理大脑活动，以解释涉及注意力、学习、记忆、决策和语言的关系。这种多学科的努力还将使培训新一代工程师和临床医生成为可能，特别强调代表性不足的群体，他们可以使用新的数据分析来帮助他们在决策期间评估大脑的健康状况，从而实现更早的诊断。这项研究的结果将对医疗保健产生重大影响，并使美国经济和社会受益。为了捕捉高度复杂的时空大脑活动，这笔赠款支持开发一种基于模型的方法，捕捉大脑动态的非平稳性和分形行为。因此，它将揭示动态特征，这些特征可以用来定量和定性地衡量一个人在特定任务(例如注意力/记忆力)中做得如何。此外，本项目中的基于模型的方法制定了执行控制的框架，其中可以使用量化措施来建立关于执行给定任务的成功的控制目标。与目前使用的短程记忆模型不同，该项目专注于一个通用的数学框架，该框架捕捉大脑活动的记忆程度。因此，这项研究将提供关于大脑组织和认知的新见解和信息，可能从根本上改变人们检查神经相关数据的方式。这一奖项反映了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。

Understanding the brain in the context of its interactions with the world is key to assessing its dynamics in health and disease states. Functional magnetic resonance imaging (fMRI) technology is able to track blood-oxygen-level dependent signals over time and serve as a proxy to neural activity. Current approaches have limitations, as they assume that the interdependence between distinct brain regions is constant throughout recording periods. These approaches also assume that the brain is an isolated system that does not consider outside stimulus. To capture the highly complex dynamics and to enable the understanding of the neural basis of human cognition this research project will develop new data analysis methods for processing brain activity to explain the relationships involved in attention, learning, memory, decision-making, and language. This multidisciplinary effort will also enable the training of a new generation of engineers and clinicians, with particular emphasis on underrepresented groups, who can use the new data analytics to aid them during decision-making to assess the brain’s health enabling earlier diagnostics. The results of this research will have a great impact on healthcare and benefit the U.S. economy and society. To capture the highly complex spatiotemporal brain activity, this grant supports development of a model-based approach that captures the non-stationarity and fractal behavior of the brain dynamics. As a consequence, it will unveil dynamical characteristics that can be used to quantitatively and qualitatively measure how well an individual is doing in a particular task (e.g., attention/memory). Furthermore, the model-based approach in this project lays down the framework to perform control where quantitative measures can be used to establish control objectives regarding success in performing a given task. Unlike currently employed short-range memory models, this project focuses on a generalized mathematical framework that captures the degree of a memory of brain activity. Thus, the research will provide new insights and information about the brain organization and cognition that may fundamentally change the way one examines neuro-related data.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.