Provenance in neuroimaging

Abstract

Provenance, the description of the history of a set of data, has grown more important with the proliferation of research consortia-related efforts in neuroimaging. Knowledge about the origin and history of an image is crucial for establishing data and results quality; detailed information about how it was processed, including the specific software routines and operating systems that were used, is necessary for proper interpretation, high fidelity replication and re-use. We have drafted a mechanism for describing provenance in a simple and easy to use environment, alleviating the burden of documentation from the user while still providing a rich description of an image's provenance. This combination of ease of use and highly descriptive metadata should greatly facilitate the collection of provenance and subsequent sharing of data.

Figure 1

Provenance Editor. A screen capture of the Provenance Editor demonstrating a subset of the information extracted from a DICOM header file and a number of user modifiable tags.

Figure 2

Pipeline Module Definition. A screen capture of a LONI Pipeline module definition for the Alignlinear (AIR 5.2.5) executable.

Figure 3

An Independent Components Analysis Workflow. A screen capture of a LONI Pipeline workflow for the calculation of the independent components of a series of BOLD images.

Figure 4

Different Architectures Yield Different Results. A, The result of an independent components analysis of a series of BOLD functional MRI brain images using workflow A. The lower right panel shows the intensity distribution of the image. The scale is 10⁵ voxels. B, The difference between images A and C rendered in false color to highlight differences. The lower right panel shows the intensity distribution of the image. The scale is 10⁵ voxels. C, The result of an independent components analysis of a series of BOLD functional MRI brain images using workflow B. The lower right panel shows the intensity distribution of the image. The scale is 10⁵ voxels.

Figure 5

A Non-linear Alignment Workflow. A screen capture of a LONI Pipeline workflow for the non-linear alignment of two images.

Figure 6

Different Compilation Options Yield Different Results. A, A T2-weigthed mouse brain image that has been aligned to a minimum deformation atlas using workflow C. The lower right panel shows the intensity distribution of the image. The scale is 10⁵ voxels. B, The difference between panels A and C rendered in false color to highlight differences. The lower right panel shows the intensity distribution of the image. The scale is 10⁵ voxels. C, A T2-weigthed mouse brain image that has been aligned to a minimum deformation atlas using workflow D. The lower right panel shows the intensity distribution of the image. The scale is 10⁵ voxels.

Figure 7

Complex Heterogeneous Pipeline. A screen capture of a complex, heterogeneous LONI Pipeline workflow demonstrating that any number of executables can be described in the Pipeline and therefore represented by the provenance schema.