Similar protocols

To access compelling stats and trends, optimize your time and resources and pinpoint new correlations, you will need to subscribe to our premium service.


Pipeline publication

[…] b"rocessing of diffusion MRI images in the research community. The main procedure comprises three parts: (I) pre-processing, (II) production of diffusion metrics, and III) construction of brain networks. Part I includes the following steps: (1) converting DICOM files into NIfTI images, (2) estimating the brain mask, (3) cropping raw images to reduce the memory cost and accelerate processing in subsequent steps, (4) correcting for the eddy-current effect, (5) averaging multiple acquisitions, and (6) calculating diffusion metrics. Part II consists of normalizing and computing multi-level diffusion metrics that can be directly used for voxel-level, atlas-level and tract-based spatial statistics (TBSS) level statistical analysis. Part III tasks include defining network nodes (i.e., parcellating gray matter into multiple regions) and constructing brain networks using deterministic and probabilistic tractography, respectively. Overall, the entire workflow of PANDA comprises 176 post-processing jobs for a single diffusion MRI dataset., The post-processing jobs within PANDA are organized using PSOM. The dependencies between jobs are first defined. According to PSOM, all jobs that are independent of all other jobs can be executed in parallel, including post-processing jobs for different individuals and independent jobs for the same subject. The processing status of each subject can be viewed in the GUI, with each job being assigned a status of \xe2\x80\x9cwait,\xe2\x80\x9d \xe2\x80\x9csubmitted,\xe2\x80\x9d \xe2\x80\x9crunning,\xe2\x80\x9d \xe2\x80\x9cfailed.\xe2\x80\x9d or \xe2\x80\x9cfinished.\xe2\x80\x9d, In PANDA, a results folder named \xe2\x80\x9cquality control\xe2\x80\x9d is generated. Snapshot pictures of the gray-matter parcellation atlas and of the fractional anisotropy and T1 images in both native and standard spaces are saved to this folder. These pictures can be used to quickly confirm the quality of the signal-noise ratio of the raw image, the quality of the spatial normalization, and the quality of the gray-matter parcellation. For the construction of the brain network, snapshot pictures of the whole-brain white-matter-tract map, which is derived from whole-brain deterministic tractography, are also produced to confirm quality by visual inspection., Image post-processing procedures within PANDA are carefully implemented, and frequently used parameters are set as the default. PANDA was thoroughly tested and validated by students and collaborators before its official release., Flexible (e.g., LONI Pipeline, JIST, and Nipype) and fixed (e.g., CIVET, PANDA, and DPARSF) parallel workflow tools are both widely used in the neuroimaging field. These tools can substantially simplify human brain MRI post-processing and optimize available computational resources. The tool choice for a specific study depends on application domains, users' background/preference, and access to computational resources., Using flexible workflow tools, users can create any desired workflow using available modules in the library together with user-generated modules. To establish an appropriate workflow in such a modular environment, users need to have a good understanding of each individual module as well as the entire workflo" […]

Pipeline specifications

Software tools FSL, PANDA, JIST