Macro- and microvascular biases in laminar BOLD fMRI: insights from realistic vascular modeling
Poster No:
1905
Submission Type:
Abstract Submission
Authors:
Grant Hartung1,2, Jonathan Polimeni2,3
Institutions:
1Technical University of Darmstadt, Marburg, Germany, 2Athinoula A. Martinos Center for Biomedical Imaging / MGH / Harvard Medical School, Charlestown, MA, 3Massachusetts Institute of Technology, Cambridge, MA
First Author:
Grant Hartung, PhD
Technical University of Darmstadt|Athinoula A. Martinos Center for Biomedical Imaging / MGH / Harvard Medical School
Marburg, Germany|Charlestown, MA
Technical University of Darmstadt|Athinoula A. Martinos Center for Biomedical Imaging / MGH / Harvard Medical School
Marburg, Germany|Charlestown, MA
Co-Author:
Jonathan Polimeni
Athinoula A. Martinos Center for Biomedical Imaging / MGH / Harvard Medical School|Massachusetts Institute of Technology
Charlestown, MA|Cambridge, MA
Athinoula A. Martinos Center for Biomedical Imaging / MGH / Harvard Medical School|Massachusetts Institute of Technology
Charlestown, MA|Cambridge, MA
Introduction:
Laminar fMRI attempts to infer patterned neuronal activity in distinct cerebral cortical layers (Norris et al. 2019). Interpretation of BOLD fMRI signals measured across depths in cortical gray matter (GM) relies on assumptions of the vascular architecture and its impact on the BOLD signal. For example, large pial veins impose local maxima in the BOLD response, causing the BOLD response to decrease progressively away from the pial surface. Moreover, some observations of a "bump" in the middle cortical layers attribute this to stronger activation (Koopmans et al. 2010). However, capillary density varies across cortical depth (Duvernoy et al. 1981) and may impart a bias on the BOLD signal (Hartung et al. 2023). Thus, accounting for potential "microvascular biases" is needed for proper laminar BOLD fMRI data interpretation.
Recent modeling efforts include more vascular realism to account for these biases (Havlicek et al. 2020). Here, we use realistic vascular anatomical network (VAN) models, representing all intracortical blood vessels at one location, to investigate the effects of structural and hemodynamic properties on the resulting BOLD profile.
Recent modeling efforts include more vascular realism to account for these biases (Havlicek et al. 2020). Here, we use realistic vascular anatomical network (VAN) models, representing all intracortical blood vessels at one location, to investigate the effects of structural and hemodynamic properties on the resulting BOLD profile.
Methods:
We used VANs reconstructed from microscopy data from the mouse primary somatosensory region (S1) (Blinder et al. 2013) and a VAN synthesized from published human microvascular statistics (Hartung et al. 2022) (Figure 1A-B). We simulated hemodynamic (Gagnon et al. 2015) and oxygenation (Hartung et al. 2021) responses and the associated magnetic field offsets (Gagnon et al. 2015) corresponding to a 2 s forepaw stimulation (Uhlirova et al. 2016). We then grafted a large (75-μm radius) vein onto the reconstructed mouse VAN to investigate the impact on the BOLD results. we also compared the impact of baseline oxygen tension (pO2) by comparing uniform CMRO2 across depths (a first-approximation or "null hypothesis") and a varying CMRO2 that decreases with depth as suggested by recent observations (Mächler et al. 2022). We evaluated these CMRO2 models on five synthetic VANs with different vascular architecture with peak capillary density regions in: (i) upper half, (ii) "layer IV", (iii) lower half, (iv) middle third, and (v) uniform as shown Figure 2. The total blood volume and number of vessels were kept constant across these synthetic VANs.
Results:
As expected, the BOLD cortical depth profile follows that of cCBV0 (Figure 1C). To identify contributors to the "bump" in the middle depths, we compared the BOLD response from each vascular compartment individually (arteries, capillaries, and veins) and found a strong pial response in the veins and a bump in the capillary response. The nearly uniform cCBV in the human VAN model results in a BOLD profile that decreases steadily with cortical depth.
The uniform CMRO2 model leads to a steadily decreasing baseline pO2 through the cortex while the depth-varying CMRO2 model gives a flatter profile (Figure 2A-B). The uniform CMRO2 model also leads to a "notch", or a local minimum, in the BOLD profile in all VANs in Figure 2, which is absent in the depth-varying CMRO2 model.
The addition of a large pial vein creates considerable changes in the local magnetic field (Figure 2C), although it only creates a notable increase in the BOLD profile near the pial surface (<250 µm depth).
The uniform CMRO2 model leads to a steadily decreasing baseline pO2 through the cortex while the depth-varying CMRO2 model gives a flatter profile (Figure 2A-B). The uniform CMRO2 model also leads to a "notch", or a local minimum, in the BOLD profile in all VANs in Figure 2, which is absent in the depth-varying CMRO2 model.
The addition of a large pial vein creates considerable changes in the local magnetic field (Figure 2C), although it only creates a notable increase in the BOLD profile near the pial surface (<250 µm depth).
Conclusions:
We demonstrated that large pial veins only affect the BOLD profile near the pial surface, as expected as the magnetic fields decrease quadratically with distance from vessels. The baseline pO2 also appears to impact the BOLD profile. While our capillary density variations are larger than expected in vivo, we observed a clear bias in the BOLD profile due to changes in capillary CBV0 and baseline pO2. Our predictive model incorporates measured data (i.e., microvascular anatomy) and, using first-principle biophysics (e.g., mass conservation), tests how specific anatomical and physiological features influence the BOLD response.
Modeling and Analysis Methods:
Activation (eg. BOLD task-fMRI)
Novel Imaging Acquisition Methods:
BOLD fMRI 1
Physiology, Metabolism and Neurotransmission:
Cerebral Metabolism and Hemodynamics 2
Keywords:
Cerebral Blood Flow
fMRI CONTRAST MECHANISMS
FUNCTIONAL MRI
HIGH FIELD MR
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.
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:
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):
Was this research conducted in the United States?
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.
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.
Please indicate which methods were used in your research:
For human MRI, what field strength scanner do you use?
Provide references using APA citation style.
Gagnon, Louis et al. 2015. “Quantifying the Microvascular Origin of BOLD-fMRI from First Principles with Two-Photon Microscopy and an Oxygen-Sensitive Nanoprobe.” Journal of Neuroscience 35 (8): 3663–75. https://doi.org/10.1523/JNEUROSCI.3555-14.2015.
Hartung, Grant et al. 2021. “Voxelized Simulation of Cerebral Oxygen Perfusion Elucidates Hypoxia in Aged Mouse Cortex.” PLoS Computational Biology 17 (1): e1008584.
———. 2022. “Simulated fMRI Responses Using Human Vascular Anatomical Network Models with Varying Architecture and Dynamics.” In . Vol. 30. London, UK.
———. 2023. “Capillary Density Induces ‘Microvascular Biases’ in BOLD: Insights from Realistic Vascular Modeling.” In . Vol. 29. Toronto, CAN.
Havlicek, Martin et al. 2020. “A Dynamical Model of the Laminar BOLD Response.” NeuroImage 204:116209.
Koopmans, Peter J. et al. 2010. “Layer‐specific BOLD Activation in Human V1.” Human Brain Mapping 31 (9): 1297–1304. https://doi.org/10.1002/hbm.20936.
Mächler, Philipp et al. 2022. “Baseline Oxygen Consumption Decreases with Cortical Depth.” PLoS Biology 20 (10): e3001440.
Norris, David G et al. 2019. “Laminar (f) MRI: A Short History and Future Prospects.” Neuroimage 197:643.
Uhlirova, Hana et al. 2016. “Cell Type Specificity of Neurovascular Coupling in Cerebral Cortex.” Elife 5:e14315.
Acknowledgments
This work was supported in part by the NIH NIBIB (grants P41-EB030006 and R01-EB032746), NIH NINDS (R01-NS128843), by the BRAIN Initiative (NIH NINDS grant U19-NS123717 and NIH NIBIB, R01-EB033206), by the German Deutsche Forschungsgemeinschaft (DFG Project number 543670971), and by the MGH/HST Athinoula A. Martinos Center for Biomedical Imaging. Computational resources were generously provided by the Massachusetts Life Sciences Center (https://www.masslifesciences.com/).