|
Size: 2198
Comment:
|
Size: 3641
Comment:
|
| Deletions are marked like this. | Additions are marked like this. |
| Line 4: | Line 4: |
| = Samseg = | = Samseg (cross-sectional, longitudinal, MS lesions) = |
| Line 6: | Line 6: |
| == Introduction == | '''''This functionality is available in [[https://surfer.nmr.mgh.harvard.edu/fswiki/ReleaseNotes|FreeSurfer 7]], with gradual improvements appearing in the development version.''''' |
| Line 8: | Line 8: |
| '''SAMseg''' is the general name of the processing stream intended to replace the subcortical segmentation stream in FreeSurfer, including some of the intensity correction and skull stripping steps preceding mri_ca_register/label. | ''Author: Koen Van Leemput'' |
| Line 10: | Line 10: |
| SAMseg is an acronym for Sequence-Adaptive Model for Segmentation, and is based on the work of Oula Puonti, Juan Eugenio Iglesias and Koen Van Leemput: [[http://orbit.dtu.dk/files/127427974/Fast_and_sequence_adaptive.pdf|Fast and Sequence-Adaptive Whole-Brain Segmentation Using Parametric Bayesian Modeling]] | ''E-mail: koen [at] nmr.mgh.harvard.edu'' |
| Line 12: | Line 12: |
| Work on Samseg is driven by three distinct grant aims: | ''Rather than directly contacting the author, please post your questions on this module to the FreeSurfer mailing list at freesurfer [at] nmr.mgh.harvard.edu'' |
| Line 14: | Line 14: |
| 1. Doug Greve, FreeSurfer Maintenance R01: Replacement of the current FreeSurfer segmentation stream with one that is faster, and most importantly accepts multimodal data. This grant is also funding acquisition of subject data to compose a new multi-modal, labeled atlas, in addition to algorithm development. | If you use these tools in your analysis, please cite: |
| Line 16: | Line 16: |
| 2. CorticoMetrics, Morphometry Phase II SBIR: Replacement of the current FS seg stream with one that is faster, but still based solely on T1-weighted input, and using the existing FS atlas. As such, it is a technical milestone of the Maintenance grant, but one conducted by CorticoMetrics in its grant with the sole focus on speed of execution relative to current freesurfer. | * Cross-sectional: [[http://nmr.mgh.harvard.edu/~koen/PuontiNI2016.pdf|Fast and sequence-adaptive whole-brain segmentation using parametric Bayesian modeling]]. O. Puonti, J.E. Iglesias, K. Van Leemput. Neuroimage, 143, 235-249, 2016. |
| Line 18: | Line 18: |
| 3. Andre van der Kouwe, FS on-the-scanner (future submission): The ability to generate structure segmentations in near-real-time on an MRI scanner (or GPU'd extension), possibly accepting lower-res data and less-than current FS quality segmentations (in the name of identifying FOV of structures to acquire). | * Longitudinal: [[https://arxiv.org/pdf/2008.05117.pdf|A Longitudinal Method for Simultaneous Whole-Brain and Lesion Segmentation in Multiple Sclerosis]]. S. Cerri, A. Hoopes, D.N. Greve, M. Mühlau, K. Van Leemput. International Workshop on Machine Learning in Neuroimaging, 2020. |
| Line 20: | Line 20: |
| This page is intended to document the multiple fronts of work composing these aims. | * MS lesions: [[https://arxiv.org/pdf/2005.05135|A Contrast-Adaptive Method for Simultaneous Whole-Brain and Lesion Segmentation in Multiple Sclerosis]]. S. Cerri, O. Puonti, D.S. Meier, J. Wuerfel, M. Mühlau, H.R. Siebner, K. Van Leemput. 2020. |
| Line 22: | Line 22: |
| == Project Fronts == | See also: ThalamicNuclei, HippocampalSubfieldsAndNucleiOfAmygdala, BrainstemSubstructures |
| Line 24: | Line 24: |
| * Evaluate existing Samseg tools - See [[SamsegEvaluationMarch2017]] * Bring Samseg tools into FS repository * Replacement of SPM registration tool - See [[SamsegAffine]] * Testing of Samseg-T1 against FS aseg - See [[SamsegAsegTesting]] * Subject acquisition |
<<BR>> === 1. General Description === Sequence Adaptive Multimodal SEGmentation (SAMSEG) is a tool to robustly segment dozens of brain structures from head MRI scans without preprocessing. The characteristic property of SAMSEG is that it accepts multi-contrast MRI data without prior assumptions on the specific type of scanner or pulse sequences used. Dedicated versions to handle longitudinal data, or to segment white matter lesions in multiple sclerosis (MS) patients are also available. The figure below illustrates a typical SAMSEG segmentation result on a T1w-FLAIR scan of a MS patient: <<BR>> {{attachment:3D_small.png||height="500"}} <<BR>><<BR>> |
| Line 31: | Line 34: |
| === 2. Basic SAMSEG (cross-sectional processing) === | |
| Line 32: | Line 36: |
| In its most basic form SAMSEG takes one or more co-registered MRI volumes as input, and produces an output segmentation in around 10 min on a good desktop computer (with multi-threading enabled). Preprocessing of the scan(s) with FreeSurfer is neither required nor recommended (e.g., no reformatting to 1mm isotropic resolution, no bias field correction and no skull stripping is needed nor recommended). The command line is: | |
| Line 33: | Line 38: |
| {{{ run_samseg --input <file> [<file> ...] --output <dir> [--threads <threads>] [--pallidum-separate] }}} where: * ''<file>'': is the path to the input volume(s) in NIFTI or MGZ file format. If you have more than one contrast (e.g., both T1w and T2w) you can simply list all the input contrasts you want to use -- the only requirement is that all input volumes are co-registered with each other, and have the same image grid size and voxel dimensions (see below for instructions on how to do that with FreeSurfer). * ''<dir>'': is the path to the output directory where the results will be saved. If this directory doesn't exist, it will be created automatically. * ''<threads> (optional)'': is the number of threads to be used by SAMSEG (default uses one thread). Set the number of threads to the number of CPU cores on your computer to get the fastest run time. * <<BR>> === 3. Usage === |
This page is readable only by those in the LcnGroup and CmetGroup.
Samseg (cross-sectional, longitudinal, MS lesions)
This functionality is available in FreeSurfer 7, with gradual improvements appearing in the development version.
Author: Koen Van Leemput
E-mail: koen [at] nmr.mgh.harvard.edu
Rather than directly contacting the author, please post your questions on this module to the FreeSurfer mailing list at freesurfer [at] nmr.mgh.harvard.edu
If you use these tools in your analysis, please cite:
Cross-sectional: Fast and sequence-adaptive whole-brain segmentation using parametric Bayesian modeling. O. Puonti, J.E. Iglesias, K. Van Leemput. Neuroimage, 143, 235-249, 2016.
Longitudinal: A Longitudinal Method for Simultaneous Whole-Brain and Lesion Segmentation in Multiple Sclerosis. S. Cerri, A. Hoopes, D.N. Greve, M. Mühlau, K. Van Leemput. International Workshop on Machine Learning in Neuroimaging, 2020.
MS lesions: A Contrast-Adaptive Method for Simultaneous Whole-Brain and Lesion Segmentation in Multiple Sclerosis. S. Cerri, O. Puonti, D.S. Meier, J. Wuerfel, M. Mühlau, H.R. Siebner, K. Van Leemput. 2020.
See also: ThalamicNuclei, HippocampalSubfieldsAndNucleiOfAmygdala, BrainstemSubstructures
1. General Description
Sequence Adaptive Multimodal SEGmentation (SAMSEG) is a tool to robustly segment dozens of brain structures from head MRI scans without preprocessing. The characteristic property of SAMSEG is that it accepts multi-contrast MRI data without prior assumptions on the specific type of scanner or pulse sequences used. Dedicated versions to handle longitudinal data, or to segment white matter lesions in multiple sclerosis (MS) patients are also available.
The figure below illustrates a typical SAMSEG segmentation result on a T1w-FLAIR scan of a MS patient:
2. Basic SAMSEG (cross-sectional processing)
In its most basic form SAMSEG takes one or more co-registered MRI volumes as input, and produces an output segmentation in around 10 min on a good desktop computer (with multi-threading enabled). Preprocessing of the scan(s) with FreeSurfer is neither required nor recommended (e.g., no reformatting to 1mm isotropic resolution, no bias field correction and no skull stripping is needed nor recommended). The command line is:
run_samseg --input <file> [<file> ...] --output <dir> [--threads <threads>] [--pallidum-separate]
where:
<file>: is the path to the input volume(s) in NIFTI or MGZ file format. If you have more than one contrast (e.g., both T1w and T2w) you can simply list all the input contrasts you want to use -- the only requirement is that all input volumes are co-registered with each other, and have the same image grid size and voxel dimensions (see below for instructions on how to do that with FreeSurfer).
<dir>: is the path to the output directory where the results will be saved. If this directory doesn't exist, it will be created automatically.
<threads> (optional): is the number of threads to be used by SAMSEG (default uses one thread). Set the number of threads to the number of CPU cores on your computer to get the fastest run time.