Data sharing and publishing in the field of neuroimaging

One crucial issue is how producers of shared data can and should be acknowledged and how this important component of science will benefit individuals in their academic careers. While we encourage the field to make use of these opportunities for data publishing, it is critical that standards for metadata, provenance, and other descriptors are used. This commentary outlines the efforts of the International Neuroinformatics Coordinating Facility Task Force on Neuroimaging Datasharing to coordinate and establish such standards, as well as potential ways forward to relieve the issues that researchers who produce these massive, reusable community resources face when making the data rapidly and freely available to the public. Both the technical and human aspects of data sharing must be addressed if we are to go forward.With the worldwide push for more open science and data sharing [1], it is an ideal time to consider the current state of data sharing in neuroscience, and in particular neuroimaging research. A huge amount of neuroimaging data has been acquired around the world; a recent literature search on PubMed led to an estimate of 12 000 datasets or 144 000 scans (around 55 petabytes of data) over the past 10？years, but only a few percent of such data is available in public repositories. Over the past two years, the International Neuroinformatics Coordinating Facility (http://www.incf.org webcite) has investigated barriers to data sharing through task force working groups and public workshops, and has identified a number of roadblocks, many of which are readily addressable, that impede researchers from both sharing and making use of existing shared data. These include a lack of simple tools for finding, uploading, and downloading shared data; uncertainty about how to best organize and prepare data for sharing, and concerns about data attribution. Many researchers are also wary of data sharing because of confusion institutional human research subject protection and the