CRCNS: Geometry-based Brain Connectome Analysis

成像技术已经取得了显著的进步，在许多人类研究中得到了常规和广泛的应用，可以无创地测量人类大脑的结构和功能。扩散磁共振成像（dMRI）和结构磁共振成像（sMRI）被用来推断数百万相互连接的白质纤维束的位置，这些白质纤维束被称为大脑连接体，它们是神经活动和大脑交流的高速公路。越来越多的证据表明，一个人的大脑连接组在认知功能、行为、发展心理健康和神经精神疾病的风险方面起着重要作用。对脑连接体结构与表型和暴露之间关系的机制理解的改进有可能彻底改变精神健康障碍的预防和治疗。然而，图像采集与连接体构建和数据分析之间的巨大差距限制了进展。该项目开发了一个变革性的数据处理和分析方法工具箱，以便更好地构建、表示和分析人类大脑连接体。这些工具将应用于人类连接组项目和英国生物银行数据集，以加强对大脑连接组如何根据个体特征和暴露以及神经精神疾病而变化的理解。该工具箱将经过严格验证，包括基于扫描扫描数据的再现性和判别能力评估、样本外预测性能、模拟研究中的功率和I型错误率，以及结果的机制可解释性。具体目标有四个：(1)对连接体进行几何重构，以减少测量误差，提高鲁棒性、可重复性和判别能力；(2)连接体的几何表征以新颖的方式表征连接体，比典型的邻接矩阵表示（依赖于预先指定的感兴趣区域之间的连接强度的单一测量）编码更多的信息；(3)通过新的多尺度模型和算法将连接体与人类特征联系起来，提高了将脑连接体与表型（认知功能、行为、心理健康状况）、暴露（物质使用）和协变量（年龄、性别）联系起来的统计分析的能力和机制洞察力；

There have been remarkable advances in imaging technology, used routinely and pervasively in many human studies, that non-invasively measures human brain structure and function. Diffusion magnetic resonance imaging (dMRI) and structural MRI (sMRI) are used to infer locations of millions of interconnected white matter fiber tracts-known as the brain connectome-that act as highways for neural activity and communication across the brain. Evidence is increasing that an individual's brain connectome plays a fundamental role in cognitive functioning, behavior, and the risk of developing mental health and neuropsychiatric disorders. Improved mechanistic understanding of relationships between brain connectome structure and phenotypes and exposures has the potential to revolutionize prevention and treatment of mental health disorders. However, large gaps between the state of the art in image acquisition and in connectome construction and data analysis have limited progress. This project develops a transformative toolbox of data processing and analysis methods for better construction, representation, and analysis of human brain connectomes. These tools will be applied to the Human Connectome Project and UK Biobank datasets, to enhance understanding of how the brain connectome varies according to individual traits and exposures and with neuropsychiatric conditions. The toolbox will be rigorously validated, including assessments of reproducibility and discriminative ability based on scan-rescan data, out-of-sample predictive performance, power and type I error rates in simulation studies, and mechanistic interpretability of the results. There are four Specific Aims: (1) Geometric reconstruction of connectomes to reduce measurement errors and enhance robustness, reproducibility and discriminative power; (2) Geometric representation of connectomes characterizing connectomes in novel ways to encode much more information than is available in typical adjacency matrix representations that rely on a single measure of connection strength between pre-specified regions of interest; (3) Relating connectomes to human traits through new multiscale models and algorithms that improve power and mechanistic insight in statistical analyses relating brain connectomes to phenotypes (cognitive functioning, behavior, mental health conditions), exposures (substance use), and covariates (age, gender); (4) Dissemination of publicly available, well-documented software for routine implementation of the proposed toolbox.