In vivo MRI histology revisited: en route to hMRI toolbox 1.0
Poster No:
1832
Submission Type:
Abstract Submission
Authors:
Baris Ugurcan1, Christophe Phillips2, Luke Edwards1, Antoine Lutti3, Martina Callaghan4, Siawoosh Mohammadi5, Laurent Lamalle6, Pierre-Louis Bazin7, Bogdan Draganski8, Gunther Helms9, Karsten Tabelow10, Nikolaus Weiskopf1,11,12
Institutions:
1Department of Neurophysics, Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig, Germany, 2Université de Liège, Liège, ., 3Department of Clinical Neuroscience, University Hospital of Lausanne and University of Lausanne, Lausanne, Switzerland, 4Dept. of Imaging Neuroscience, UCL Queen Square Institute of Neurology, University College London, London, UK, 5Department of Neuroradiology, Lübeck University, Lübeck, Germany, 6Université de Liège, Liège, Belgium, 7Full brain picture Analytics, Leiden, Netherlands, 8Department of Neurology, Inselspital, Bern University Hospital, University of Bern, Bern, Switzerlan, Bern, Bern, 9Department of Clinical Sciences, Lund University, Lund, Sweden, 10Weierstraß-Institut für Angewandte Analysis und Stochastik (WIAS), Berlin, Germany, 11Felix Bloch Institute for Solid State Physics, Faculty of Physics and Earth System Sciences, Leipzig University, Leipzig, Germany, 12Wellcome Centre for Human Neuroimaging, Institute of Neurology, University College London, London, United Kingdom
First Author:
Baris Ugurcan
Department of Neurophysics, Max Planck Institute for Human Cognitive and Brain Sciences
Leipzig, Germany
Department of Neurophysics, Max Planck Institute for Human Cognitive and Brain Sciences
Leipzig, Germany
Co-Author(s):
Luke Edwards
Department of Neurophysics, Max Planck Institute for Human Cognitive and Brain Sciences
Leipzig, Germany
Department of Neurophysics, Max Planck Institute for Human Cognitive and Brain Sciences
Leipzig, Germany
Antoine Lutti
Department of Clinical Neuroscience, University Hospital of Lausanne and University of Lausanne
Lausanne, Switzerland
Department of Clinical Neuroscience, University Hospital of Lausanne and University of Lausanne
Lausanne, Switzerland
Martina Callaghan
Dept. of Imaging Neuroscience, UCL Queen Square Institute of Neurology, University College London
London, UK
Dept. of Imaging Neuroscience, UCL Queen Square Institute of Neurology, University College London
London, UK
Bogdan Draganski
Department of Neurology, Inselspital, Bern University Hospital, University of Bern, Bern, Switzerlan
Bern, Bern
Department of Neurology, Inselspital, Bern University Hospital, University of Bern, Bern, Switzerlan
Bern, Bern
Nikolaus Weiskopf
Department of Neurophysics, Max Planck Institute for Human Cognitive and Brain Sciences|Felix Bloch Institute for Solid State Physics, Faculty of Physics and Earth System Sciences, Leipzig University|Wellcome Centre for Human Neuroimaging, Institute of Neurology, University College London
Leipzig, Germany|Leipzig, Germany|London, United Kingdom
Department of Neurophysics, Max Planck Institute for Human Cognitive and Brain Sciences|Felix Bloch Institute for Solid State Physics, Faculty of Physics and Earth System Sciences, Leipzig University|Wellcome Centre for Human Neuroimaging, Institute of Neurology, University College London
Leipzig, Germany|Leipzig, Germany|London, United Kingdom
Introduction:
The hMRI toolbox is an internationally developed SPM-based framework for creation and statistical analysis of quantitative MRI maps sensitive to myelin and iron content (https://hmri.info; Tabelow et al. 2019, Draganski et al. 2011). The toolbox team uses Github to manage the collaborative software development workflow (Fig. 1) which includes: collecting user queries through issues and mailing list, organized coding and test development structured by pull requests, careful code review and testing of the new features before getting merged to the master branch. Since its initial release (v0.2) in 2018, the toolbox has acquired several new functionalities. A standalone version of the toolbox not requiring Matlab has also been released and included in the Neurodesk platform. Here we present these new features, which are part of the planned (for early 2025) major public release 1.0.
·FIGURE 1: hMRI toolbox software development flow
Methods:
The new features that we would like to highlight are:
1. Compilation of the toolbox and inclusion in Neurodesk (https://www.neurodesk.org/). The standalone version can be run directly without a Matlab license. It can either be used through the SPM batch GUI or through SPM batch jobs submitted directly on the command line.
2. Denoising module (Edwards et al., 2024). We have wrapped publicly available implementations of the state-of-the-art denoising methods (Bazin et al. 2019, Does et al. 2016, Veraart et al., 2019) within our toolbox through our own GUI, pre-processing modules and a Java-Matlab interface. Our implementation also possesses meta-data capabilities which outputs JSON sidecar files for further processing.
3. Error Maps (Mohammadi et al., 2022) give the voxelwise error of the quantitative maps for quality assurance, which can be turned on by the parameter hmri_def.errormaps in hmri defaults. An additional submodule ('Combine two successive hmri datasets') can also use these error maps as weights to robustly combine quantitative maps from two successive acquisitions.
4. Analysis of QUantitative Imaging data using a Quality Index (QUIQI) module for motion-robust group analysis (Lutti et al., 2022). The module implements a data-driven solution to account for subject specific motion, by assigning weights to each map based on an index of image quality, when performing group-level GLM estimations.
1. Compilation of the toolbox and inclusion in Neurodesk (https://www.neurodesk.org/). The standalone version can be run directly without a Matlab license. It can either be used through the SPM batch GUI or through SPM batch jobs submitted directly on the command line.
2. Denoising module (Edwards et al., 2024). We have wrapped publicly available implementations of the state-of-the-art denoising methods (Bazin et al. 2019, Does et al. 2016, Veraart et al., 2019) within our toolbox through our own GUI, pre-processing modules and a Java-Matlab interface. Our implementation also possesses meta-data capabilities which outputs JSON sidecar files for further processing.
3. Error Maps (Mohammadi et al., 2022) give the voxelwise error of the quantitative maps for quality assurance, which can be turned on by the parameter hmri_def.errormaps in hmri defaults. An additional submodule ('Combine two successive hmri datasets') can also use these error maps as weights to robustly combine quantitative maps from two successive acquisitions.
4. Analysis of QUantitative Imaging data using a Quality Index (QUIQI) module for motion-robust group analysis (Lutti et al., 2022). The module implements a data-driven solution to account for subject specific motion, by assigning weights to each map based on an index of image quality, when performing group-level GLM estimations.
Results:
The results (Fig. 2) corresponding to the new features are:
1. The standalone version (also included in Neurodesk) increases the reach of the toolbox and contributes to cross platform efforts. The main functionalities can be used by installing the freely available "Matlab runtime environment" without any need for a Matlab license.
2. The built-in denoising module can be seamlessly configured and piped with other main processing steps through the SPM batch dependencies. LCPCA-denoising, originally written in Java, can now be invoked also within/from Matlab through an advanced custom developed Java-Matlab interface, which is a reflection of the cross-platform and advanced software initiatives of the toolbox.
3. Error maps enable evaluation of both local data quality variations and artifacts without requiring additional data. The resulting robust MPM parameters show reduced variability at the group level compared to the single-repeat or averaged counterparts.
4. The QUIQI module mitigates the effects of head motion on group comparison/regression statistical analysis.
1. The standalone version (also included in Neurodesk) increases the reach of the toolbox and contributes to cross platform efforts. The main functionalities can be used by installing the freely available "Matlab runtime environment" without any need for a Matlab license.
2. The built-in denoising module can be seamlessly configured and piped with other main processing steps through the SPM batch dependencies. LCPCA-denoising, originally written in Java, can now be invoked also within/from Matlab through an advanced custom developed Java-Matlab interface, which is a reflection of the cross-platform and advanced software initiatives of the toolbox.
3. Error maps enable evaluation of both local data quality variations and artifacts without requiring additional data. The resulting robust MPM parameters show reduced variability at the group level compared to the single-repeat or averaged counterparts.
4. The QUIQI module mitigates the effects of head motion on group comparison/regression statistical analysis.
·FIGURE 2: QUIQI, Denoising module and Error maps example results
Conclusions:
The new developments demonstrate the toolbox's dedication to cross platform efforts (inclusion of the compiled-standalone toolbox in Neurodesk), advanced software techniques (custom developed Java-Matlab interface and user focused GUI developments) and high quality image processing modules offering built-in denoising, error quantification, and motion mitigation. The hMRI toolbox 1.0 is prepared and developed as a self-sufficient, easily configurable, cross platform and well-documented software, which will further support developments in MRI-based in vivo histology.
Modeling and Analysis Methods:
Other Methods
Neuroinformatics and Data Sharing:
Informatics Other 1
Novel Imaging Acquisition Methods:
Anatomical MRI
Multi-Modal Imaging 2
Imaging Methods Other
Keywords:
MRI
MRI PHYSICS
Open-Source Code
Open-Source Software
STRUCTURAL MRI
Other - Quantitative MRI
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?
Which processing packages did you use for your study?
Provide references using APA citation style.
Callaghan et al. (2019). Example dataset for the hMRI toolbox. Data in brief, 25, 104132.
Does et al. (2019). Evaluation of principal component analysis image denoising on multi‐exponential MRI relaxometry. Magnetic resonance in medicine, 81(6), 3503-3514.
Draganski et al. (2011). Regional specificity of MRI contrast parameter changes in normal ageing revealed by voxel-based quantification (VBQ). Neuroimage, 55(4), 1423-1434.
Edwards et al. (2024). Denoising improves contrast while retaining sharpness of high resolution multiparameter R1, R2* and proton density maps. ESMRMB 2024 2-5 October Barcelona: 40th Annual Scientific Meeting.
Friston et al. (2002). Classical and Bayesian inference in neuroimaging: theory. NeuroImage, 16(2), 465-483.
Lutti et al. (2022). Restoring statistical validity in group analyses of motion‐corrupted MRI data. Human Brain Mapping, 43(6), 1973-1983.
Mohammadi et al. (2022). Error quantification in multi-parameter mapping facilitates robust estimation and enhanced group level sensitivity. NeuroImage, 262, 119529.
Tabelow et al.(2019). hMRI–A toolbox for quantitative MRI in neuroscience and clinical research. Neuroimage, 194, 191-210.
Veraart et al. (2016). Denoising of diffusion MRI using random matrix theory. Neuroimage, 142, 394-406.