Neuroinformatics for Prospective Data Management and Sharing

描述（由申请人提供）：美国国立卫生研究院资助的研究人员非常需要能够轻松分享和传播脑成像研究数据的工具。获取大型神经成像数据集是了解大脑在健康和疾病中如何工作的关键方面。最典型的是，大型数据集由一名调查人员收集，用于特定目的，然后通常存档。有时，这些数据与合作者共享，通常是在长期努力之后。这种方法对从这些丰富的数据集中可以学到的东西的看法非常狭隘。例如，通常有多个问题可以问，但在研究设计时没有考虑到。还有许多先进的分析技术和方法，由多个小组开发，可以应用于这些数据，以识别隐藏的结构或回答新的问题。然而，多模态神经影像数据的数据共享存在许多障碍，多年来医学界一直在努力克服这些障碍。在这个第一阶段的STTR中，我们提出了一种独特的方法，即采用我们在心智研究网络开发的神经信息学工具，创建一个用户友好的神经信息学套件，使研究、评估和神经成像数据的前瞻性管理和共享成为可能。我们认为，处理包括隐私方面在内的前瞻性（正在进行的）研究的能力对于鼓励分享至关重要，因为是否与谁分享的控制权将在研究者个人的控制之下。这样，共享就非常容易实现，因为所有数据都是在一个更大的基础设施中收集、存储和管理的。我们计划探索我们方法的可行性，这将是实现和维持我们的神经信息学工具所需的市场服务。这将包括安装、支持、培训和数据存储等。这个项目的成功完成将代表着一个重大的转变，它将推动不同类型数据的科学共享和知识提取成为一个实用和广泛使用的模型。2

DESCRIPTION (provided by applicant): There is a tremendous need for tools that allow NIH sponsored investigators to easily share and disseminate data from brain imaging studies. Access to large neuroimaging data sets is a critical aspect of learning about how the brain works both in health and disease. Most typically, large data sets are collected by one investigator, used for a specific purpose, and then often archived. Sometimes these data are shared with collaborators, typically after an extended effort. Such an approach takes a very narrow view of what could be learned from these rich data sets. For example, there are often multiple questions that could be asked but weren't conceived at the time of the study design. There are also a plethora of advanced analytic techniques and approaches, developed by multiple groups, which can be applied to these data to identify hidden structure or answer new questions. However, there are many barriers to data sharing of multimodal neuroimaging data and the community has tried for years to confront the various barriers. In this Phase I STTR we are proposing a unique approach which is to take the neuroinformatics tools that we have been developing at the Mind Research Network and create a user friendly neuroinformatics suite which will enable prospective management and sharing of studies, assessments, and neuroimaging data. The ability to handle prospective (ongoing) studies including privacy aspects is, we believe, critical to encouraging sharing as the control of whether and who to share with will be under the individual investigator's control. Sharing is then quite simple to enable as all the data are collected, stored, and managed within a larger infrastructure. We plan to explore the feasibility of our approach, which will be to market the services required to implement and sustain our neuroinformatics tools. This will include installation, support, training, and data storage among other things. The successful completion of this project would represent a major transformation which could propel scientific sharing and knowledge extraction of diverse types of data into a practical and widely used model. 2