Poster No:
253
Submission Type:
Late-Breaking Abstract Submission
Authors:
Alma Socorro Torres-Torres1, Jelle Dalenberg2, Remco Renken2, Michael Biehl3, Marina de Koning-Tijssen2
Institutions:
1University Medical Center Groningen, Groningen, NM, 2University Medical Center Groningen, Groningen, Groningen, 3University of Groningen, Groningen, Groningen
First Author:
Co-Author(s):
Remco Renken
University Medical Center Groningen
Groningen, Groningen
Introduction:
ET is a prevalent movement disorder, affecting approximately 1.3% of the global population [1]. Characterized by involuntary oscillatory postural tremors in the upper limbs, ET significantly impacts daily activities and reduces quality of life. Structural MRI studies suggest widespread gray and white matter alterations; however, the specific structural patterns contributing to ET remain unclear [2]. Employing interpretable classification models, such as GMLVQ-driven feature selection, may enhance biomarker discovery for ET.
This study aims to optimize the identification of structural MRI biomarkers in Essential Tremor (ET) by applying Generalized Matrix Learning Vector Quantization (GMLVQ) for classification against healthy volunteers (HV) using ROI features derived from probability maps to identify the most informative ones.
Methods:
Twenty ET patients from the Next Move in Movement Disorders (NEMO) project [3] and twenty age-matched healthy volunteers (HV) were included. T1-weighted MRI scans were preprocessed as described in [4] and further processed with FreeSurfer (v7.4.1) to obtain gray matter probability (GMP) and white matter probability (WMP) maps.
For each subject, GMP and WMP maps were used to extract regions of interest (ROIs) based on the AAL3 atlas. From each ROI, four voxel distribution features-mean, standard deviation (STD), kurtosis, and skewness-were extracted, serving as input features for classification.
To enhance ROI feature selection, we integrated recursive feature elimination (RFE) with Stratified K-Fold cross-validation (k=5). Using GMP and WMP maps, we investigated structural MRI alterations distinguishing ET from HV, ensuring feature robustness through permutation testing. RFE with GMLVQ was implemented, iteratively removing the least informative voxel clusters based on feature relevance extracted from GMLVQ's Omega matrix. After model convergence, permutation importance was applied to validate feature stability across folds. Statistical significance was set at p < 0.001, FDR-corrected (α = 0.05).
Results:
The RFE with GMLVQ achieved high classification performance (AUC = 0.91 for GMP, AUC = 0.86 for WMP). The most informative GMP ROIs distinguishing ET from HV were the left pulvinar thalamic nucleus, right cerebellum (III-IX), and left cerebellum (IV-V). For WMP, the most discriminative ROIs were the left pulvinar thalamic nucleus, right cerebellum (III-IX), left cerebellum (III-V), and right posterior cingulate gyrus.
Conclusions:
Our findings demonstrate the effectiveness of RFE with GMLVQ in identifying discriminative structural MRI biomarkers for ET classification. The pulvinar thalamic nucleus and cerebellum emerged as key differentiating regions [5-8], reinforcing the role of the cerebello-thalamo-cortical network in ET pathophysiology.
Disorders of the Nervous System:
Neurodegenerative/ Late Life (eg. Parkinson’s, Alzheimer’s) 1
Modeling and Analysis Methods:
Classification and Predictive Modeling 2
Multivariate Approaches
Segmentation and Parcellation
Keywords:
Cerebellum
Machine Learning
Movement Disorder
1|2Indicates the priority used for review
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.
I accept
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:
I am submitting this abstract as an original work to be reproduced. I am available to be the “source party” in an upcoming team and consent to have this work listed on the OSSIG website. I agree to be contacted by OSSIG regarding the challenge and may share data used in this abstract with another team.
Please indicate below if your study was a "resting state" or "task-activation” study.
Other
Healthy subjects only or patients (note that patient studies may also involve healthy subjects):
Patients
Was this research conducted in the United States?
No
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.
Yes
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.
Not applicable
Please indicate which methods were used in your research:
Structural MRI
For human MRI, what field strength scanner do you use?
1T
Which processing packages did you use for your study?
Free Surfer
Provide references using APA citation style.
1. E. D. Louis and M. McCreary, “How Common is Essential Tremor? Update on the Worldwide Prevalence of Essential Tremor,” Tremor and Other Hyperkinetic Movements, vol. 11, no. 1, 2021.
2. Luo, R., Pan, P., Xu, Y., & Chen, L. No reliable gray matter changes in essential tremor. Neurological Sciences. 2019 https://doi.org/10.1007/s10072-019-03933-0
3. A. M. M. van der Stouwe et al., “The next move in movement disorders (NEMO): developing a computer aided classification tool for hyperkinetic movement disorders.,” BMJ Open, 2021.
4. J. R. Dalenberg et al., “Next move in movement disorders: neuroimaging protocols for hyperkinetic movement disorders,” Front. Hum. Neurosci., vol. 18, p. 1406786, Aug. 2024.
5. Cameron, E., Dyke, J., Hernandez, N., Louis, E., & Dydak, U. Cerebral gray matter volume losses in essential tremor: A case-control study using high resolution tissue probability maps.. Parkinsonism & related disorders. 2018; 51. https://doi.org/10.1016/j.parkreldis.2018.03.008
6. Benito‐León, J., Alvarez-Linera, J., Hernandez-Tamames, J., Alonso-Navarro, H., Jiménez-Jiménez, F., & Louis, E. Brain structural changes in essential tremor: Voxel-based morphometry at 3-Tesla. Journal of the Neurological Sciences. 2009; 287. https://doi.org/10.1016/j.jns.2009.08.037
7. Younger, E., Ellis, E., Parsons, N., Pantano, P., Tommasin, S., Caeyenberghs, K., Benito‐León, J., Romero, J., Joutsa, J., & Corp, D. Mapping Essential Tremor to a Common Brain Network Using Functional Connectivity Analysis. Neurology. 2023; 101. https://doi.org/10.1212/WNL.0000000000207701
8. Pietracupa, S., Bologna, M., Tommasin, S., Berardelli, A., & Pantano, P. The Contribution of Neuroimaging to the Understanding of Essential Tremor Pathophysiology: a Systematic Review. The Cerebellum. 2021; 21. https://doi.org/10.1007/s12311-021-01335-7
No