Layer-sensitive fMRI

Poster No:

1210 

Submission Type:

Abstract Submission 

Authors:

Gal Hershkovitz1, Omri Tomer1, Ittai Shamir1, Daniel Barazany1, Yaniv Assaf1

Institutions:

1Tel Aviv University, Tel Aviv, Israel

First Author:

Gal Hershkovitz  
Tel Aviv University
Tel Aviv, Israel

Co-Author(s):

Omri Tomer  
Tel Aviv University
Tel Aviv, Israel
Ittai Shamir  
Tel Aviv University
Tel Aviv, Israel
Daniel Barazany, Dr.  
Tel Aviv University
Tel Aviv, Israel
Yaniv Assaf  
Tel Aviv University
Tel Aviv, Israel

Introduction:

The distribution of cortical layers across the brain has a major role in activation patterns in response to stimuli. Even though layers' width is too small for being visualized when using conventional fMRI (~1mm), recent studies have shown that, while using high-field, high-resolution fMRI, it's possible to distinguish between deep and superficial layers (Huber et al., 2017). Other studies have also shown that it's possible to anatomically discriminate the cortical layers based on their T1 values (Barazany and Assaf, 2011; Lifshits et al., 2018; Shamir et al.,2019). Here we suggest to use low-resolution inversion recovery filtered fMRI to nullify, or considerably reduce the BOLD signal that originates from a tissue component whose T1 is associated with specific cortical layer. Motor and sensory tasks, being known to be different in layer-specific activation patterns (Weiler et al., 2008; Mountcastle, 1957), were used to validate our layer-sensitive fMRI approach.

Methods:

MRI was performed on a Siemens 3T Magnetom Prisma. Nineteen participants were scanned with a series of Inversion-Recovery filtered Spin-Echo (IR-SE) fMRI block designs aiming to discriminate between M1 and S1 by motor (fist-clenching) and sensory (hand brushing by the experimenter) tasks. Layer sensitive BOLD response was achieved by varying the inversion time (TI) between 630 and 750 ms (focused on gray matter T1) with TE/TR of 28/3000 ms. Subjects were additionally scanned without TI filtering, with TE/TR of 30/1500 ms. Each fMRI design consisted of 4 blocks of 15 seconds, for a total of 135 seconds.
Functional activation maps of M1 and S1 response to stimuli were computed from unfiltered SE experiment on the group level (validated by comparison with the Juelich histological atlas). The activation ROIs were then projected back to the subject space to calculate the mean and peak BOLD response for M1 and S1 for both tasks under each TI filtering.
Finally, we created maps where each activated voxel was colored according to the TI where its peak BOLD response was most significantly reduced. It is expected that the TI where minimum BOLD response is observed in the IR filtered fMRI will represent the layer of activation.

Results:

Unfiltered fMRI 1st and 2nd level analysis' statistics maps were created t>2.8 threshold, and showed significant activation during motor and sensory tasks in M1 and S1 accordingly.
M1 and S1 ROIs were located by relying on the statistics maps, and transformed into subjects' space. When examining these ROIs by extracting mean BOLD response and its peak for each IR-filtered fMRI and each task, we found a significant difference in nullification patterns of motor and sensory activations (F = 13.87, p<0.003).
Fig1 shows that for the motor task the minimum BOLD response was achieved at TI of 690ms and for the sensory task it was around 660ms, indicating that most significant nullification of BOLD signal originated from different T1-related layers.
For whole brain visualization of layer-sensitive nullification we've defined Time to Nullify (TN) as the TI that most significantly reduced voxels' BOLD response, and generated subject-space TN maps for each activation mask.
As can be seen in Fig2, there are clear differences between T1-related layers' nullification patterns in response to stimuli. Signal originating from motor regions during motor task was nullified mostly by longer TIs (~750ms), in contrast to sensory regions, which during sensory tasks were mostly nullified by shorter TIs (~690ms).
Supporting Image: Picture1.jpg
Supporting Image: Picture3.jpg
 

Conclusions:

We've shown in this work that by altering the TI during IR-filtered fMRI, we were able to visualize T1-related layer nullification patterns, that differs between different stimuli. Layers' T1 variation across brain regions is heavily affected by their myeloarchitectonic structure, and is the basis of the approach presented here. Although our method is currently subject-specific, it paves the way for further investigation of layer-specific functionality.

Modeling and Analysis Methods:

Activation (eg. BOLD task-fMRI)
Connectivity (eg. functional, effective, structural)
Methods Development 1
Segmentation and Parcellation

Novel Imaging Acquisition Methods:

BOLD fMRI 2

Keywords:

Atlasing
Computational Neuroscience
Cortex
Cortical Layers
FUNCTIONAL MRI
Modeling
Motor
MRI PHYSICS
Segmentation
Somatosensory

1|2Indicates the priority used for review

My abstract is being submitted as a Software Demonstration.

No

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 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

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.

Not applicable

Please indicate which methods were used in your research:

Functional MRI
Structural MRI

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

3.0T

Which processing packages did you use for your study?

FSL

Provide references using author date format

Huber, L., Handwerker, D. A., Jangraw, D. C., Chen, G., Hall, A., Stüber, C., ... & Goense, J. (2017). High-resolution CBV-fMRI allows mapping of laminar activity and connectivity of cortical input and output in human M1. Neuron, 96(6), 1253-1263.
Barazany, D., & Assaf, Y. (2011). Visualization of cortical lamination patterns with magnetic resonance imaging. Cerebral Cortex, 22(9), 2016-2023.
Lifshits, S., Tomer, O., Shamir, I., Barazany, D., Tsarfaty, G., Rosset, S., & Assaf, Y. (2018). Resolution considerations in imaging of the cortical layers. NeuroImage, 164, 112-120.
Shamir, I., Tomer, O., Baratz, Z., Tsarfaty, G., Faraggi, M., Horowitz, A., & Assaf, Y. (2019). A framework for cortical laminar composition analysis using low-resolution T1 MRI images. Brain Structure and Function, 224(4), 1457-1467.
Weiler, N., Wood, L., Yu, J., Solla, S. A., & Shepherd, G. M. (2008). Top-down laminar organization of the excitatory network in motor cortex. Nature neuroscience, 11(3), 360.
Mountcastle, V. B. (1957). Modality and topographic properties of single neurons of cat's somatic sensory cortex. Journal of neurophysiology, 20(4), 408-434.