TemplateFlow

Reference anatomies of the brain and corresponding atlases play a central role in experimental neuroimaging workflows and are the foundation for reporting standardized results. The choice of such references —i.e., templates— and atlases is one relevant source of methodological variability across studies, which has recently been brought to attention as an important challenge to reproducibility in neuroscience. TemplateFlow is a publicly available framework for human and nonhuman brain models. The framework combines an open database with software for access, management, and vetting, allowing scientists to distribute their resources under FAIR —findable, accessible, interoperable, reusable— principles. TemplateFlow supports a multifaceted insight into brains across species, and enables multiverse analyses testing whether results generalize across standard references, scales, and in the long term, species, thereby contributing to increasing the reliability of neuroimaging results.

参考大脑解剖结构和相应的图谱在实验神经影像学工作流程中占据核心地位，并构成了报告标准化结果的基础。此类参考——即模板——及图谱的选择，是研究间方法论差异性的一个重要来源，这一问题近年来已引起关注，被视为神经科学中可重复性的重要挑战。TemplateFlow是一个公开可用的框架，旨在为人类和非人类大脑模型提供支持。该框架结合了一个开放数据库以及用于访问、管理和审核的软件，使得科学家能够在FAIR——即易于发现、易于访问、互操作性、可重复使用——的原则下分发其资源。TemplateFlow支持对跨物种大脑的多维度洞察，并允许进行多元宇宙分析，以检验结果是否能够推广至标准参考、尺度，以及在长期内，物种，从而有助于提高神经影像学结果的可靠性。