The Human Connectome Phantom (HCPh) dataset
Poster No:
2233
Submission Type:
Abstract Submission
Authors:
Oscar Esteban1, Céline Provins2, Elodie Savary3, Alexandre Cionca4, Hélène Lajous5, Yasser Alemán-Gómez6, Eleonora Fornari7, Benedetta Franceschiello8, Ileana Jelescu9, Patric Hagmann6
Institutions:
1Lausanne University Hospital and University of Lausanne, Lausanne, VD, 2Lausanne University Hospital and University of Lausanne, Lausanne, Vaud, 3Lausanne University Hospital and University of Lausanne, Lausanne, -, 4Centre Hospitalier Universitaire Vaudoise, Lausanne, Vaud, 5Lausanne University Hospital and University of Lausanne, Lausanne, Switzerland, 6Department of Radiology, Lausanne University Hospital and University of Lausanne (CHUV-UNIL), Lausanne, Switzerland, 7CHUV, Lausanne/Switzerland, Switzerland, 8The Sense Innovation and Research Centre, Sion, Switzerland, 9Lausanne University Hospital (CHUV), Lausanne, Vaud
First Author:
Co-Author(s):
Yasser Alemán-Gómez
Department of Radiology, Lausanne University Hospital and University of Lausanne (CHUV-UNIL)
Lausanne, Switzerland
Department of Radiology, Lausanne University Hospital and University of Lausanne (CHUV-UNIL)
Lausanne, Switzerland
Patric Hagmann
Department of Radiology, Lausanne University Hospital and University of Lausanne (CHUV-UNIL)
Lausanne, Switzerland
Department of Radiology, Lausanne University Hospital and University of Lausanne (CHUV-UNIL)
Lausanne, Switzerland
Introduction:
Network-based approaches are widely adopted to model functional and structural 'connectivity' of the living brain, extracted noninvasively with magnetic resonance imaging (MRI). However, these analyses -on functional and structural networks- render unreliable at the finer temporal, spatial, and brain-parcellation scales. Here, we introduce the Human Connectome PHantom (HCPh) dataset as part of our recent Stage 1 Registered Report¹ (Fig. 1).
Methods:
Data. One healthy participant (author OE), left-handed, aged 40 at the onset of the study underwent 50 sessions (14 of which were piloting sessions or sessions excluded for insufficient quality of the data or circumstantial impediments to finalizing the protocol) with a comprehensive MRI protocol on a single, 3T Siemens PrismaFit scanner. The protocol includes anatomical (T₁-weighted, T₂-weighted), high-angular-resolution multi-shell diffusion-weighted imaging, three blood-oxygen-level-dependent (BOLD) functional MRI runs (including a quality control task, QCT, of 2 m 38 s, a breath-holding task, BHT, of 5 m 41 s, and a naturalistic timelapse watching akin to resting-state of 20min 3s). The MRI protocol was acquired while simultaneously recording eye-tracking, electrocardiogram (ECG), respiratory motion with a pneumatic belt, and gas concentration collection through a nasal cannula (O₂ and CO₂). A second wave of data is scheduled to start the collection of 12 sessions on each of three different 3T scanners using a single protocol (36 additional sessions).
Standard Operating Procedures (SOPs). All the data collection and methodological implementation details are comprehensively specified in the SOPs document², which is openly available at www.axonlab.org/hcph-sops under a CC-BY license.
Data management plan. Data are converted into BIDS³ and version-controlled with DataLad⁴ as described in the SOPs. Quality assessment and control of unprocessed data will be implemented with MRIQC⁵. Data will be preprocessed with NeuroImaging PREProcessing toolS (NiPreps; www.nipreps.org) -particularly fMRIPrep⁶ and dMRIPrep⁷. The original BIDS dataset, as well as all derivatives, will be openly shared upon culmination of Stage 2 of the Registered Report.
Standard Operating Procedures (SOPs). All the data collection and methodological implementation details are comprehensively specified in the SOPs document², which is openly available at www.axonlab.org/hcph-sops under a CC-BY license.
Data management plan. Data are converted into BIDS³ and version-controlled with DataLad⁴ as described in the SOPs. Quality assessment and control of unprocessed data will be implemented with MRIQC⁵. Data will be preprocessed with NeuroImaging PREProcessing toolS (NiPreps; www.nipreps.org) -particularly fMRIPrep⁶ and dMRIPrep⁷. The original BIDS dataset, as well as all derivatives, will be openly shared upon culmination of Stage 2 of the Registered Report.
·Figure 1
·Table 1
Results:
A dataset to address hypotheses about the instruments as opposed to the brain. By very densely collecting data on a single individual, we aim to investigate questions about the MRI imaging device and interactions with physiological and instrumental sources of variability. While this dataset will not provide any new insights into behavior or white matter microstructure, it is the first dataset to 'calibrate' functional and structural connectivity neuroimaging pipelines, enabling their comparison and isolating the different sources of variability throughout the workflow. Table 1 (reproduced from the corresponding Registered Report) highlights the potential contribution of the HCPh in the landscape of open or at least reportedly accessible dense MRI datasets.
A comprehensive, version-controlled experimental reporting. Our SOPs have been created using the NiPreps' 'SOPs-cookiecutter' project (nipreps.org/sops-cookiecutter/), which maximizes the shareability of the SOPs documentation, keeps version control leveraging Git and may employ GitHub's features for code review, automated actions (e.g., to check spelling errors or normalize the style of the Markdown code), and publishing while protecting sensitive information (e.g., usernames and passwords) by a design that keeps secrets inaccessible from the public repository.
A resource to harmonize data across scanners. With the collection of a wealth of multi-scanner data (n=12 sessions per scanner), this dataset will help develop new harmonization techniques without introducing individual differences in models.
A comprehensive, version-controlled experimental reporting. Our SOPs have been created using the NiPreps' 'SOPs-cookiecutter' project (nipreps.org/sops-cookiecutter/), which maximizes the shareability of the SOPs documentation, keeps version control leveraging Git and may employ GitHub's features for code review, automated actions (e.g., to check spelling errors or normalize the style of the Markdown code), and publishing while protecting sensitive information (e.g., usernames and passwords) by a design that keeps secrets inaccessible from the public repository.
A resource to harmonize data across scanners. With the collection of a wealth of multi-scanner data (n=12 sessions per scanner), this dataset will help develop new harmonization techniques without introducing individual differences in models.
Conclusions:
The HCPh dataset will be openly released to be used as calibration data on network analyses, as exploratory data, or for educational purposes.
Modeling and Analysis Methods:
Connectivity (eg. functional, effective, structural) 2
Neuroinformatics and Data Sharing:
Databasing and Data Sharing 1
Novel Imaging Acquisition Methods:
Anatomical MRI
BOLD fMRI
Diffusion MRI
Keywords:
ADULTS
Computational Neuroscience
Data Organization
Design and Analysis
FUNCTIONAL MRI
Informatics
Open-Source Code
Pre-registration
STRUCTURAL MRI
WHITE MATTER IMAGING - DTI, HARDI, DSI, ETC
1|2Indicates the priority used for review
Provide references using author date format
2. Provins, C. et al. The Human Connectome PHantom (HCPh) study: Standard Operating Procedures. Online Rep. (2023) doi:10.5281/zenodo.8383184.
3. Poldrack, R. A. et al. The Past, Present, and Future of the Brain Imaging Data Structure (BIDS). Preprint at https://doi.org/10.48550/arXiv.2309.05768 (2023).
4. Halchenko, Y. O. et al. DataLad: distributed system for joint management of code, data, and their relationship. J. Open Source Softw. 6, 3262 (2021).
5. Esteban, O. et al. MRIQC: Advancing the automatic prediction of image quality in MRI from unseen sites. PLOS ONE 12, e0184661 (2017).
6. Esteban, O. et al. fMRIPrep: a robust preprocessing pipeline for functional MRI. Nat. Methods 16, 111–116 (2019).
7. Joseph, M. J. E. et al. dMRIPrep: a robust preprocessing pipeline for diffusion MRI. in Proc. Intl. Soc. Mag. Reson. Med. vol. 30 2473 (2021).