The brainlife.io cloud-services for functional network neuroscience
Poster No:
1674
Submission Type:
Abstract Submission
Authors:
Joshua Faskowitz1, Conner Victory2, David Hunt1, Franco Delogu2, Soichi Hayashi1, Richard Betzel1, Franco Pestilli1
Institutions:
1Indiana University, Bloomington, IN, 2Lawrence Technological University, Southfield, MI
First Author:
Co-Author(s):
Introduction:
Using functional magnetic resonance imaging (fMRI), we can measure the brain's distributed functional organization. Maps of fMRI activity can be used to create functional networks, which in turn can be analyzed using the tools of network science to uncover brain-wide properties such as functional community organization [10] or hub-like structure [6].
The field of Network Neuroscience exists at the intersection of human brain mappers and network science practitioners. These fields require both advanced software skills and mathematical knowledge. Whereas on the one hand, fMRI specialists learn to employ highly specialized image processing techniques and must make sure their data is artifact-free, on the other hand, network scientists focus on learning and developing innovative network science algorithms applicable across fields. To achieve expert-level knowledge in both domains is both a challenge and a barrier for investigators and trainees in either field.
Our work promotes FAIR principles [9] by addressing the challenges highlighted above. We present a series of cloud computing services that make network neuroscience more accessible by enabling the generation of functional brain networks in a streamlined and intuitive manner. The services comprise of containerized "Apps" that process MRI data from the raw NIFTI files (for both fMRI and T1-weighted anatomy) to node-by-node functional connectivity matrices. These services can be run automatically on the various datasets available on OpenNeuro.org, BIDS data hoster on DataLad.org, or on user-uploaded data via a point-and-click web-interface. The interface allows users to take advantage of a powerful distributed cloud computing infrastructure via brainlife.io. Finally, brainlife.io generates a full provenance record for the data generated by keeping track of Apps (and versions) used to build the brain networks, supporting the aim of computational reproducibility [5].
The field of Network Neuroscience exists at the intersection of human brain mappers and network science practitioners. These fields require both advanced software skills and mathematical knowledge. Whereas on the one hand, fMRI specialists learn to employ highly specialized image processing techniques and must make sure their data is artifact-free, on the other hand, network scientists focus on learning and developing innovative network science algorithms applicable across fields. To achieve expert-level knowledge in both domains is both a challenge and a barrier for investigators and trainees in either field.
Our work promotes FAIR principles [9] by addressing the challenges highlighted above. We present a series of cloud computing services that make network neuroscience more accessible by enabling the generation of functional brain networks in a streamlined and intuitive manner. The services comprise of containerized "Apps" that process MRI data from the raw NIFTI files (for both fMRI and T1-weighted anatomy) to node-by-node functional connectivity matrices. These services can be run automatically on the various datasets available on OpenNeuro.org, BIDS data hoster on DataLad.org, or on user-uploaded data via a point-and-click web-interface. The interface allows users to take advantage of a powerful distributed cloud computing infrastructure via brainlife.io. Finally, brainlife.io generates a full provenance record for the data generated by keeping track of Apps (and versions) used to build the brain networks, supporting the aim of computational reproducibility [5].
Methods:
To build a functional brain network, users needs (1) anatomical (T1-weighted MR) and (2) functional (fMR) image for each subject. Anatomical images are used to compute a parcellation [4], which serves as nodes in a functional brain network. The services are compatible with several of the currently available atlases, e.g., Schaefer [8], Yeo [10]) via the Multi-Atlas Transfer Tool App (https://doi.org/10.25663/bl.app.23) and FreeSurfer (https://doi.org/10.25663/bl.app.0). fMRI data preprocessing is implemented via the fMRIPrep App (https://doi.org/10.25663/brainlife.app.160) [3]. Finally, given a completed fMRIPrep output and anatomical parcellation, functional connectivity matrices are built by regressing-out noise parameters (https://doi.org/10.25663/brainlife.app.167) [7].
Results:
The Apps described above are publicly available (on Github) and integrated within brainlife.io, Figure 1a,b shows the User Interface with the Apps icons and subject-level data. Figure 1c displays the provenance graph for a single data object (a functional network). Data and Apps related to this abstract can be accessed at https://doi.org/10.25663/brainlife.pub.10.
Conclusions:
Constructing functional brain networks requires specialized knowledge, creating a barrier to entry for network neuroscientists and a dearth of available data for network scientists. We address this barrier by building cloud-computing tools to construct these networks. Our future work will focus on: 1) adding fMRI quality control feedback [2] and 2) adding network analysis algorithms to Brainlife.io, so that full analytical pipelines, from raw data to p-value, can be run in a reproducible manner on the cloud.
Modeling and Analysis Methods:
Connectivity (eg. functional, effective, structural)
fMRI Connectivity and Network Modeling 1
Neuroinformatics and Data Sharing:
Databasing and Data Sharing
Workflows 2
Keywords:
FUNCTIONAL MRI
Informatics
1|2Indicates the priority used for review
My abstract is being submitted as a Software Demonstration.
Please indicate below if your study was a "resting state" or "task-activation” study.
Healthy subjects only or patients (note that patient studies may also involve healthy subjects):
Are you Internal Review Board (IRB) certified? Please note: Failure to have IRB, if applicable will lead to automatic rejection of abstract.
Was any human subjects research approved by the relevant Institutional Review Board or ethics panel? NOTE: Any human subjects studies without IRB approval will be automatically rejected.
Was any animal research approved by the relevant IACUC or other animal research panel? NOTE: Any animal studies without IACUC approval will be automatically rejected.
Please indicate which methods were used in your research:
For human MRI, what field strength scanner do you use?
Provide references using author date format
[2] Esteban, O., (2017). MRIQC: Advancing the automatic prediction of image quality in MRI from unseen sites. PloS one, 12(9), e0184661.
[3] Esteban, O., (2019). fMRIPrep: a robust preprocessing pipeline for functional MRI. Nature methods, 16(1), 111.
[4] Fischl, B. (2012). FreeSurfer. Neuroimage, 62(2), 774-781.
[5] Poldrack, (2017). Scanning the horizon: towards transparent and reproducible neuroimaging research. Nature Reviews Neuroscience, 18(2), 115.
[6] Power, J. D., Schlaggar, B. L., Lessov-Schlaggar, C. N., & Petersen, S. E. (2013). Evidence for hubs in human functional brain networks. Neuron, 79(4), 798-813.
[7] Satterthwaite, T. D., (2013). An improved framework for confound regression and filtering for control of motion artifact in the preprocessing of resting-state functional connectivity data. Neuroimage, 64, 240-256.
[8] Schaefer, A., (2017). Local-global parcellation of the human cerebral cortex from intrinsic functional connectivity MRI. Cerebral Cortex, 28(9), 3095-3114.
[9] Wilkinson, M. D., (2016). The FAIR Guiding Principles for scientific data management and stewardship. Scientific data, 3.
[10] Yeo, B. T. T, (2011). The organization of the human cerebral cortex estimated by intrinsic functional connectivity. Journal of neurophysiology, 106(3), 1125-1165.
This research was supported by NSF OAC-1916518, NSF IIS-1912270, NSF, IIS-1636893, NSF BCS-1734853, NIH 1R01EB029272-01, Google Cloud, a Microsoft Research Award, A Microsoft Investigator Fellowship, the Indiana University Areas of Emergent Research initiative “Learning: Brains, Machines, Children.” This material is based on the work supported by the National Science Foundation Graduate Research Fellowship under Grant No.1342962 (J.F.). Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.