Print  Close

Variability in CVR and tSNR Across MRI Centres: Insights from Normalised and Raw BOLD Data

Poster No:

1524 

Submission Type:

Abstract Submission 

Authors:

Ruwan Wanni Arachchige1,2, Jessikah Fildes1, Yidian Gao3,4, Iman Idrees5, Waheeda Hawa5, Aliza Finch3,4, Hannah Lyons6, Tara Ghafari3,4, Alice Waitt3,4,5, Dan Ford1, Asha Strom3,7, Matthew Brookes1, Jan Novak5, Martin Wilson3,4, Andrew Bagshaw3,4, Lisa Hill6, James Mitchell6, Samuel Lucas3,7,8, Alexandra Sinclair6, Davinia Fernandez-Espejo3,4,8, Karen Mullinger1,3,4,8, on behalf of the mTBI Predict consortium1,3,4,5,6,7

Institutions:

1Sir Peter Mansfield Imaging Centre, School of Physics and Astronomy,University of Nottingham, Nottingham, United Kingdom, 2Medical Imaging, Department of Health and Care Professions, Faculty of Health and Life Sciences, University of Exeter, Exeter, United Kingdom, 3Centre for Human Brain Health, University of Birmingham, Birmingham, United Kingdom, 4School of Psychology, University of Birmingham, Birmingham, United Kingdom, 5Institute of Health and Neurodevelopment, Aston University, Birmingham, United Kingdom, 6Translational Brain Science, College of Medicine and Health, University of Birmingham, Birmingham, United Kingdom, 7School of Sport, Exercise and Rehabilitation Sciences, University of Birmingham, Birmingham, United Kingdom, 8Contributed equally to leading the science of this work, Birmingham, United Kingdom

First Author:

Pradeepa Ruwan Wanni Arachchige  
Sir Peter Mansfield Imaging Centre, School of Physics and Astronomy,University of Nottingham|Medical Imaging, Department of Health and Care Professions, Faculty of Health and Life Sciences, University of Exeter
Nottingham, United Kingdom|Exeter, United Kingdom

Co-Author(s):

Jessikah Fildes  
Sir Peter Mansfield Imaging Centre, School of Physics and Astronomy,University of Nottingham
Nottingham, United Kingdom
Yidian Gao  
Centre for Human Brain Health, University of Birmingham|School of Psychology, University of Birmingham
Birmingham, United Kingdom|Birmingham, United Kingdom
Iman Idrees  
Institute of Health and Neurodevelopment, Aston University
Birmingham, United Kingdom
Waheeda Hawa  
Institute of Health and Neurodevelopment, Aston University
Birmingham, United Kingdom
Aliza Finch  
Centre for Human Brain Health, University of Birmingham|School of Psychology, University of Birmingham
Birmingham, United Kingdom|Birmingham, United Kingdom
Hannah Lyons  
Translational Brain Science, College of Medicine and Health, University of Birmingham
Birmingham, United Kingdom
Tara Ghafari  
Centre for Human Brain Health, University of Birmingham|School of Psychology, University of Birmingham
Birmingham, United Kingdom|Birmingham, United Kingdom
Alice Waitt  
Centre for Human Brain Health, University of Birmingham|School of Psychology, University of Birmingham|Institute of Health and Neurodevelopment, Aston University
Birmingham, United Kingdom|Birmingham, United Kingdom|Birmingham, United Kingdom
Dan Ford  
Sir Peter Mansfield Imaging Centre, School of Physics and Astronomy,University of Nottingham
Nottingham, United Kingdom
Asha Strom  
Centre for Human Brain Health, University of Birmingham|School of Sport, Exercise and Rehabilitation Sciences, University of Birmingham
Birmingham, United Kingdom|Birmingham, United Kingdom
Matthew Brookes  
Sir Peter Mansfield Imaging Centre, School of Physics and Astronomy,University of Nottingham
Nottingham, United Kingdom
Jan Novak  
Institute of Health and Neurodevelopment, Aston University
Birmingham, United Kingdom
Martin Wilson  
Centre for Human Brain Health, University of Birmingham|School of Psychology, University of Birmingham
Birmingham, United Kingdom|Birmingham, United Kingdom
Andrew Bagshaw  
Centre for Human Brain Health, University of Birmingham|School of Psychology, University of Birmingham
Birmingham, United Kingdom|Birmingham, United Kingdom
Lisa Hill  
Translational Brain Science, College of Medicine and Health, University of Birmingham
Birmingham, United Kingdom
James Mitchell  
Translational Brain Science, College of Medicine and Health, University of Birmingham
Birmingham, United Kingdom
Samuel Lucas  
Centre for Human Brain Health, University of Birmingham|School of Sport, Exercise and Rehabilitation Sciences, University of Birmingham|Contributed equally to leading the science of this work
Birmingham, United Kingdom|Birmingham, United Kingdom|Birmingham, United Kingdom
Alexandra Sinclair  
Translational Brain Science, College of Medicine and Health, University of Birmingham
Birmingham, United Kingdom
Davinia Fernandez-Espejo  
Centre for Human Brain Health, University of Birmingham|School of Psychology, University of Birmingham|Contributed equally to leading the science of this work
Birmingham, United Kingdom|Birmingham, United Kingdom|Birmingham, United Kingdom
Karen Mullinger  
Sir Peter Mansfield Imaging Centre, School of Physics and Astronomy,University of Nottingham|Centre for Human Brain Health, University of Birmingham|School of Psychology, University of Birmingham|Contributed equally to leading the science of this work
Nottingham, United Kingdom|Birmingham, United Kingdom|Birmingham, United Kingdom|Birmingham, United Kingdom
on behalf of the mTBI Predict consortium  
Sir Peter Mansfield Imaging Centre, School of Physics and Astronomy,University of Nottingham|Centre for Human Brain Health, University of Birmingham|School of Psychology, University of Birmingham|Institute of Health and Neurodevelopment, Aston University|Translational Brain Science, College of Medicine and Health, University of Birmingham|School of Sport, Exercise and Rehabilitation Sciences, University of Birmingham
Nottingham, United Kingdom|Birmingham, United Kingdom|Birmingham, United Kingdom|Birmingham, United Kingdom|Birmingham, United Kingdom|Birmingham, United Kingdom

Introduction:

Cerebrovascular reactivity (CVR) is a biomarker for assessing brain tissue health and vascular function (Xu et al., 2023) and changes with conditions including dementia, stroke and traumatic brain injury (Liu et al., 2019). Quantifying CVR using blood oxygen level-dependent (BOLD) functional MRI (fMRI) with a CO₂ vasostimulus has demonstrated utility in research and clinical applications (Liu et al., 2021). However, due to the diversity of imaging techniques, complexity of vasodilatory challenge, and data acquisition from different scanners. CVR values often exhibit substantial variability (Rovai et al., 2024; Zhao et al., 2021).In this multi-vendor study, we assessed the reproducibility of temporal signal-to-noise ratio (tSNR) and CVR metrics across sites.

Methods:

3T MRI data were collected from 19 healthy participants (mean age: 37±7.6yrs, 15 Male), using EPI-BOLD (TR=2s, TE=40ms) and T1-weighted anatomical scans across three sites: Site 1 (Philips Achieva), Sites 2 and 3 (Siemens Prisma). Participants completed six sessions: four sessions at their primary site (assignment of the primary site was equally distributed between participants) and one session at each of the other two sites. The task consisted of 2 blocks of 4-mins room air and 4-mins 5% CO2 delivered via a Douglas bag setup. End-tidal CO₂ (ETCO₂) was recorded (LabChart Lightning 1.13.2), and the partial pressure of ETCO₂ (PETCO₂) was derived based on daily barometric pressure.
MRI data were preprocessed using fMRIPrep. The mean grey matter (GM) BOLD signal was extracted and the PETCO₂ was temporally aligned to it. A general linear model (GLM) was constructed with the: PETCO₂ trace, six motion parameters, physiological noise estimates (aCompCor components from fMRIprep) (Krentz et al., 2023), and a linear drift term. β-weights for PETCO₂ regressor provided a metric of CVR. The mean PETCO₂ regressor β-weights across all GM voxels was calculated to obtain global CVR values. The GLM was run twice: 1) raw BOLD images input and 2) normalised BOLD images; where each image in the timeseries was divided by the standard deviation of the signal in the first 90s of acquisition (where no BOLD change due to CO2 challenge occurred). Additionally, tSNR maps were calculated from the first 90s of data and averaged over the GM.
Systematic differences between sites and sessions were examined by performing Pearson correlation analyses between tSNR and CVR. Repeated measures (RM)-ANOVAs were used to evaluate the site-based effects on GM CVR and tSNR measures.

Results:

Figure 1 shows that there was no significant relationship between tSNR and CVR (raw-CVR: r=0.048,p=0.620; normalised CVR: r=-0.118,p=0.222). tSNR did not significantly differ between the three sites (F (1.59,27.01) =0.25,p=0.734). However, before normalisation, RM-ANOVA revealed significant differences in CVR across the three sites (F (1.43, 24.32)=8.34,p=0.004), which was mediated by the differences between Site 1 and Sites 2&3. After normalisation, no significant differences in CVR were found across the three sites (F (1.74, 29.55)=2.13,p=0.142). Spatial maps of CVR data across sites is shown in Figure 2.
Supporting Image: Figure1.png
Supporting Image: Figure2.png
 

Conclusions:

This study shows that our optimised EPI sequence resulted in BOLD-weighted data with matched tSNR values across scanners. However, systematic differences were observed in CVR metrics between sites, in contrast to Liu et al, 2021. The CVR difference was reduced by normalising the data by the data variance before calculating CVR (Z-scoring). However, to our knowledge this is not currently standard practice such that comparison of BOLD-derived CVR measures across studies may not be valid. Future research should assess the generalisability of normalisation approaches and ensure physiological differences between patients and healthy controls are not removed by cross-site normalisation.

Acknowledgements: We thank the UK mTBI Predict Consortium. This work was funded by the Ministry of Defence, UK, through the mTBI Predict Consortium

Modeling and Analysis Methods:

Activation (eg. BOLD task-fMRI)
Methods Development 1

Physiology, Metabolism and Neurotransmission:

Cerebral Metabolism and Hemodynamics
Neurophysiology of Imaging Signals 2
Physiology, Metabolism and Neurotransmission Other

Keywords:

Cerebral Blood Flow
Cerebrovascular Disease
Data analysis
Design and Analysis
FUNCTIONAL MRI
MRI
MRI PHYSICS
NORMAL HUMAN
Trauma
Other - Cerebrovascular reactivity;normalised BOLD; temporal signal to noise ratio;multisite studies

1|2Indicates the priority used for review

Abstract Information

By submitting your proposal, you grant permission for the Organization for Human Brain Mapping (OHBM) to distribute your work in any format, including video, audio print and electronic text through OHBM OnDemand, social media channels, the OHBM website, or other electronic publications and media.

I accept

The Open Science Special Interest Group (OSSIG) is introducing a reproducibility challenge for OHBM 2025. This new initiative aims to enhance the reproducibility of scientific results and foster collaborations between labs. Teams will consist of a “source” party and a “reproducing” party, and will be evaluated on the success of their replication, the openness of the source work, and additional deliverables. Click here for more information. Propose your OHBM abstract(s) as source work for future OHBM meetings by selecting one of the following options:

I do not want to participate in the reproducibility challenge.

Please indicate below if your study was a "resting state" or "task-activation” study.

Task-activation

Healthy subjects only or patients (note that patient studies may also involve healthy subjects):

Healthy subjects

Was this research conducted in the United States?

No

Were 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.

Yes

Were any animal research approved by the relevant IACUC or other animal research panel? NOTE: Any animal studies without IACUC approval will be automatically rejected.

Not applicable

Please indicate which methods were used in your research:

Functional MRI
Computational modeling
Other, Please specify  -   Cerebrovascular reactivity

For human MRI, what field strength scanner do you use?

3.0T

Which processing packages did you use for your study?

FSL
Other, Please list  -   MATLAB

Provide references using APA citation style.

Krentz, M., Tutunji, R., Kogias, N., Mahadevan, H. M., Reppmann, Z. C., Krause, F., & Hermans, E. J. (2023). A Comparison of fMRI Data-Derived and Physiological Data-Derived Methods for Physiological Noise Correction. https://doi.org/10.1101/2023.02.22.529506

Liu, P., De Vis, J. B., & Lu, H. (2019). Cerebrovascular reactivity (CVR) MRI with CO2 challenge: A technical review. NeuroImage, 187, 104–115. https://doi.org/10.1016/j.neuroimage.2018.03.047

Liu, P., Jiang, D., Albert, M., Bauer, C. E., Caprihan, A., Gold, B. T., Greenberg, S. M., Helmer, K. G., Jann, K., Jicha, G., Rodriguez, P., Satizabal, C. L., Seshadri, S., Singh, H., Thompson, J. F., Wang, D. J. J., & Lu, H. (2021). Multi-vendor and multisite evaluation of cerebrovascular reactivity mapping using hypercapnia challenge. NeuroImage, 245. https://doi.org/10.1016/j.neuroimage.2021.118754

Rovai, A., Lolli, V., Trotta, N., Goldman, S., & De Tiège, X. (2024). CVRmap—a complete cerebrovascular reactivity mapping post-processing BIDS toolbox. Scientific Reports, 14(1). https://doi.org/10.1038/s41598-024-57572-3

Xu, B., Vu, C., Borzage, M., González-Zacarías, C., Shen, J., & Wood, J. (2023). Improved cerebrovascular reactivity mapping using coherence weighted general linear model in the frequency domain. NeuroImage, 284. https://doi.org/10.1016/j.neuroimage.2023.120448

Zhao, M. Y., Woodward, A., Fan, A. P., Chen, K. T., Yu, Y., Chen, D. Y., Moseley, M. E., & Zaharchuk, G. (2021). Reproducibility of cerebrovascular reactivity measurements: A systematic review of neuroimaging techniques *. https://doi.org/10.25740/hd852bg4538

UNESCO Institute of Statistics and World Bank Waiver Form

I attest that I currently live, work, or study in a country on the UNESCO Institute of Statistics and World Bank List of Low and Middle Income Countries list provided.

No