Mesoscopic in vivo human T 2 * dataset acquired using quantitative MRI at 7 Tesla
Curation statements for this article:-
Curated by eLife
Evaluation Summary:
The methods presented in this work are of potential broad interest across different domains of human neuroscience. Reliable methods for pushing the limits of spatial resolution for mesoscopic scale imaging of the living human cortex are of wide interest and utility. The image quality and high-spatial resolution of the data are exceptionally high. The paper in its current form demonstrates the application of the developed methods to a few exemplary cortical regions and sequences.
(This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. Reviewer #1 agreed to share their name with the authors.)
This article has been Reviewed by the following groups
Listed in
- Evaluated articles (eLife)
- Neuroscience (eLife)
Abstract
Mesoscopic (0.1-0.5 mm) interrogation of the living human brain is critical for advancing neuroscience and bridging the resolution gap with animal models. Despite the variety of MRI contrasts measured in recent years at the mesoscopic scale, in vivo quantitative imaging of T 2 * has not been performed. Here we provide a dataset containing empirical T 2 * measurements acquired at 0.35 × 0.35 × 0.35 mm 3 voxel resolution using 7 Tesla MRI. To demonstrate unique features and high quality of this dataset, we generate flat map visualizations that reveal fine-scale cortical substructures such as layers and vessels, and we report quantitative depth-dependent T 2 * (as well as R 2 *) values in primary visual cortex and auditory cortex that are highly consistent across subjects. This dataset is freely available at https://doi.org/10.17605/OSF.IO/N5BJ7 , and may prove useful for anatomical investigations of the human brain, as well as for improving our understanding of the basis of the T 2 * -weighted (f)MRI signal.
Article activity feed
-
-
-
Evaluation Summary:
The methods presented in this work are of potential broad interest across different domains of human neuroscience. Reliable methods for pushing the limits of spatial resolution for mesoscopic scale imaging of the living human cortex are of wide interest and utility. The image quality and high-spatial resolution of the data are exceptionally high. The paper in its current form demonstrates the application of the developed methods to a few exemplary cortical regions and sequences.
(This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. Reviewer #1 agreed to share their name with the authors.)
-
Reviewer #1 (Public Review):
The authors examined the information available in mesoscopic resolution structural MRI data acquired at 0.35mm isotropic resolution. Data are acquired from about a third of the brain, but the analysis concentrates on the calcarine sulcus and Heschl's gyrus. The authors locate patterns of draining veins and laminar profiles based on T1 and T2*. Based on earlier work they advocate acquiring two sets of scans with orthogonal orientations of the phase-encoding directions and then taking a minimum intensity projection to eliminate artifacts caused by pulsatile flow. In each subject they define four regions of interest and then use a novel algorithm to flatten the convoluted data.
The main strengths of the study are the high spatial resolution achieved, and the quantitative measurement of both T1 and T2*. In …
Reviewer #1 (Public Review):
The authors examined the information available in mesoscopic resolution structural MRI data acquired at 0.35mm isotropic resolution. Data are acquired from about a third of the brain, but the analysis concentrates on the calcarine sulcus and Heschl's gyrus. The authors locate patterns of draining veins and laminar profiles based on T1 and T2*. Based on earlier work they advocate acquiring two sets of scans with orthogonal orientations of the phase-encoding directions and then taking a minimum intensity projection to eliminate artifacts caused by pulsatile flow. In each subject they define four regions of interest and then use a novel algorithm to flatten the convoluted data.
The main strengths of the study are the high spatial resolution achieved, and the quantitative measurement of both T1 and T2*. In addition, the open-science approach with both data and source code are appreciated. The development of a flattening procedure that does not rely on triangular meshes is valuable.
The weaknesses are the lack of whole brain data, which may disappoint some potential users, and the lack of motion correction, which does not significantly affect the data quality of the present paper, but could cause difficulties for others wishing to implement the acquisition protocol with less compliant subjects.
A manual segmentation duration of 8-10 hours per subject is intimidating. It also seems like a missed opportunity that although multi-echo GE data were generated no attempt was made to make use of the phase information (phase maps, SWI, QSM).The authors clearly show that in V1 they can identify cortical profiles consistent with the presence of the stripe of Gennari, that is absent in Heschl's gyrus.
The paper contributes to a body of literature showing that it is possible to obtain information on myelination using both T2* and T1 parameters. It would have been interesting to see whether the current data can hint at the presence of the lines of Baillarger in the extrastriate cortex (see for example "Lines of Baillarger in vivo and ex vivo: Myelin contrast across lamina at 7 T MRI and histology, Fracasso et al. 2016"). The availability of the processing software is a valuable contribution to the community, but I would have been interested to understand how it differs from, for example, CBS tools. -
Reviewer #2 (Public Review):
This work provides tools for the acquisition and analysis of human brain MRI data at the mesoscopic scale as demonstrated in the visual and auditory cortex. Magnetic resonance imaging is a key tool for noninvasive evaluation of human brain structure and function but has been traditionally hampered by low sensitivity and spatial resolution. This paper provides acquisition strategies to surmount several barriers to achieving mesoscopic-scale MRI data, including motion secondary to blood flow in small vessels, and analysis tools to characterize changes in MRI contrast along the complex surface of the cerebral cortex. These approaches provide a framework for more robust acquisition and analysis of mesoscopic MRI data in the living human brain, particularly at ultra-high field, and serve as useful tools for …
Reviewer #2 (Public Review):
This work provides tools for the acquisition and analysis of human brain MRI data at the mesoscopic scale as demonstrated in the visual and auditory cortex. Magnetic resonance imaging is a key tool for noninvasive evaluation of human brain structure and function but has been traditionally hampered by low sensitivity and spatial resolution. This paper provides acquisition strategies to surmount several barriers to achieving mesoscopic-scale MRI data, including motion secondary to blood flow in small vessels, and analysis tools to characterize changes in MRI contrast along the complex surface of the cerebral cortex. These approaches provide a framework for more robust acquisition and analysis of mesoscopic MRI data in the living human brain, particularly at ultra-high field, and serve as useful tools for advancing human neuroscience with more detailed characterization of human brain structure and function.
Strengths:
The averaging of multi-echo gradient echo MRI data with orthogonal phase-encoding directions using minimum intensity images voxels provides a clever approach to leveraging information across multiple image acquisitions and offers a viable approach to mitigating spatial misregistration due to blood flow motion while boosting signal-to-noise ratio, which proves to be important at mesoscopic spatial resolution.
A major strength of this paper is the exquisite image quality and high spatial resolution attained in the living human cortex. The high quality of the 0.35 mm isotropic spatial resolution images with detailed segmentations of the cortex surrounding the calcarine sulcus and Heschl's gyrus, including demonstration of fine-scale cortical substructures such as the stria of Gennari and intracortical vessels, demonstrates the promise of such technology in characterizing cortical laminar architecture in the living human brain.
The comparison of T2* variations at different cortical depths in visual and auditory cortex provides a sound validation of the acquisition and analysis methods in reproducing known trends in anatomy in these different cortical regions.
The availability of the analysis tools and data through open-source software and data-sharing enables the widespread dissemination of such mesoscopic imaging data, which is difficult to acquire and not readily accessible on standard scanners.
Weaknesses:
Some of the methods demonstrated in the manuscript are not fully discussed or characterized in-depth, leading to a lack of clarity regarding how to place the technical advances in the context of existing methods. For example, the authors describe how the cortical patch flattening method has desirable distortion characteristics compared to a specific triangular mesh-based tool developed by (Kay et al., 2019), yet they do not demonstrate systematically how the local distortions induced by flattening a folded cortex may impact the representation of cortical metrics, particularly as a function of spatial resolution.
The authors claim that the acquisition and analysis methods developed in this paper represent a significant advance toward demonstrating mesoscopic scale imaging in the living human brain, yet they confine their analyses to cortical regions with well-defined differences in laminar architecture. The paper thus reads as a confirmation of largely well-known areal differences in myeloarchitecture in the human cortex, as opposed to the intended application of mesoscopic scale imaging toward uncovering subtle differences and cyto- and myeloarchitecture in areas of cortex where the laminar architectonic boundaries are less well-delineated and understood.
The paper focuses on mitigating motion artifact related to blood flow while largely glossing over the challenge of mitigating bulk head motion, which is a greater source of error for the long acquisitions required for mesoscopic-scale imaging. It would be valuable for the authors to provide more detailed information and insight into how bulk motion was mitigated in the presented data.
-