CAREER: Discovering Common Human Brain Architecture

是否存在一种共同的人类大脑结构，可以在个体之间进行定量编码和精确复制？这个CAREER项目旨在通过密集个性化和基于共同连接的皮层标志（DICCCOL）的地图来发现和表示常见的人类大脑结构。每个标志将通过从扩散张量成像（DTI）数据中导出的分组一致的白色物质纤维连接模式来定义。与此同时，将使用大规模多模式fMRI和DTI数据集来确定DICCCOL和功能定位之间的预测关系。由此产生的DICCCOL表示共同的大脑架构将被应用于创建一个通用的和个性化的大脑参考系统，构建人脑连接体，并阐明大脑的功能相互作用。该项目的教育目标是创建和评估一种全新的跨学科高等教育方法，即变革性跨学科小组学习（TIGL）。来自三门相关但来自不同学科视角的课程（生物医学图像分析，MRI物理学导论和脑功能成像）的学生和教师将在一个教室里一起工作。在这些共同的会议期间，学生将进行协同学习活动，参与跨学科小组讨论，设计和进行跨学科小组项目。共同的大脑结构的发现和表示将从根本上推进对人类大脑的科学理解。DICCCOL图谱及其预测框架的广泛传播将改变许多依赖于个体之间结构/功能对应的应用。DICCCOL地图提供了一个通用的桥梁，以比较和整合实验室之间的神经成像数据，这将刺激和实现大量的合作努力。虽然该项目的重点是脑成像，但结构和功能预测建模的一般方法预计将影响许多其他成像领域。TIGL方法将促进对跨学科学习的基本理解。TIGL方法将扩大到其他机构和学科，并将广泛传播。这种持续的努力将建立TIGL方法作为一种通用的跨学科教育方法，以提高具有跨学科思维的下一代科学家的能力。

Is there a common human brain architecture that can be quantitatively encoded and precisely reproduced across individuals? This CAREER project aims to discover and represent common human brain architecture through a map of Dense Individualized and Common Connectivity-based Cortical Landmarks (DICCCOL). Each of the landmarks will be defined by group-wise consistent white matter fiber connection patterns derived from diffusion tensor imaging (DTI) data. In parallel, large-scale multimodal fMRI and DTI datasets will be employed to determine predictive relationships between DICCCOLs and functional localizations. The resulting DICCCOL representation of common brain architecture will be applied to create a universal and individualized brain reference system, construct human brain connectomes, and elucidate the brain's functional interactions. The education objective of this CAREER project is to create and assess a fundamentally novel interdisciplinary higher education approach, namely, transformative interdisciplinary group learning (TIGL). Students and instructors from three courses that are related but emerge from different disciplinary perspectives (Biomedical Image Analysis, Introduction to MRI Physics, and Functional Brain Imaging) will work together in one classroom. During these common sessions, the students will have synergistic learning activities, engage in interdisciplinary group discussions, and design and conduct interdisciplinary group projects.The discovery and representation of common brain architecture will fundamentally advance scientific understanding of the human brain. Broad dissemination of the DICCCOL map and its prediction framework will transform numerous applications that rely on structural/functional correspondences across individuals. The DICCCOL map offers a generic bridge to compare and integrate neuroimaging data across laboratories, which will stimulate and enable plentiful collaborative efforts. While this project has a focus on brain imaging, the general methodology of predictive modeling of structure and function is expected to influence many other imaging domains. The TIGL approach will advance fundamental understanding of interdisciplinary learning. The TIGL approach will be scaled up to other institutions and disciplines, and will be widely disseminated. This continuous effort will establish the TIGL approach as a general interdisciplinary education methodology to increase the capacity of the next generation of scientists who have an interdisciplinary mindset.