Big-Data Electron-microscopy for Novel Community Hypotheses: Measuring And Retrieving Knowledge (BENCHMARK)

项目摘要 为了更好地了解神经系统的结构组织和解剖，请访问 前所未有的空间分辨率，最近的努力，包括大脑倡议资助的项目，已经 使用电子显微镜(EM)和X射线收集越来越大的数据集 显微断层摄影(XRM)。我们现在可以在一系列不同的尺度上成像神经组织， 有可能为下一代亚微米和纳米级的脑图谱奠定基础 决议。然而，数据收集方法以及正在进行的数据收集方法存在巨大的变异性 研究发展中的成像技术、实验方案、数据存储和后处理 加工方法。不同的分辨率、对比度、染色、图像校正、数据 压缩、机器学习算法和元数据都在开发中。要启用 与其他数据集和成像方式的比较、荟萃分析和配准，新 EM和XRM数据的标准是必需的，类似于光学显微镜中追求的标准， 磁共振成像等领域。在EM和XRM增长的这段时间里 成像及其在神经科学研究中越来越多的采用和利用，它是一个关键 是时候实施可确保互操作性、可持续性和可用性的标准了 昂贵的数据集。这对于实现开放、实验室之间的共享以及 在这些大而昂贵的数据集上的可重复性的结果。这项提案旨在发展 大比例尺EM和XRM结构数据标准，以及注释和 指向补充数据源的链接。这将极大地支持验证、共享和复制 在这个社区扩大对其他大脑倡议项目的投资。我们的团队将带来 将研究人员社区合并为两个互补的图像工作组(WG) 和实验元数据标准和注释标准。这一利益共同体将 与Brain计划协作制定标准并传播结果 项目和档案。最后，该项目将构建工具来查询和检索图像和 注释数据，包括Motif发现，通过社区门户和开放源码工具。 这将使科学家能够重复分析数据、测试假设和共享数据产品 并与社区一起取得成果。我们将强调与现有标准的跨领域协作 社区以及支持以下标准的软件工具的开发和集成 确保领养。

Project Summary In an effort to better understand structural organization and anatomy of nervous systems at unprecedented spatial resolution, recent efforts, including BRAIN Initiative funded projects, have collected increasingly larger datasets using Electron Microscopy (EM) and X-Ray Microtomography (XRM). We can now image neural tissue across a range of different scales, potentially forming the basis for the next generation of brain atlases at submicron and nanometer resolution. However, there is huge variability in data collection approaches, as well as ongoing research into evolving imaging technology, experimental protocols, data storage, and post- processing methods. Different resolutions, contrasts, staining, image corrections, data compression, machine learning algorithms, and metadata are all being developed. To enable comparison, meta-analysis, and registration with other datasets and imaging modalities, new standards for EM and XRM data are required, similar to those pursued in light microscopy, magnetic resonance imaging, and other domains. In this time period of growth in EM and XRM imaging, and its increased adoption and utilization for neuroscientific investigations, it is a critical time to implement standards that ensure interoperability, sustainability, and availability of these expensive datasets. This will be critical to enable openness, sharing between laboratories, and reproducible results on these large and expensive datasets. This proposal aims to develop standards for large scale EM and XRM structural data, as well as standards for annotations and links to complementary data sources. This will enable validation, sharing, and replication, greatly amplifying investment in other BRAIN initiative projects in this community. Our team will bring together a community of researchers into two complementary Working Groups (WGs) for Image and Experimental Metadata Standards and Annotation Standards. This community of interest will collaboratively develop standards and disseminate results in conjunction with BRAIN initiative projects and archives. Finally, this project will build tools to query and retrieve image and annotation data, including motif discovery, through a community portal and open source tools. This will allow scientists to reproducibly analyze data, test hypotheses, and share data products and results with the community. We will emphasize collaboration with existing standards across communities and the development and integration of software tools supporting the standards to ensure adoption.