{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,7,30]],"date-time":"2025-07-30T16:05:26Z","timestamp":1753891526466,"version":"3.41.2"},"reference-count":47,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2025,6,13]],"date-time":"2025-06-13T00:00:00Z","timestamp":1749772800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Neuroinform."],"abstract":"<jats:p>Advancements in machine learning hold great promise for the analysis of multimodal neuroimaging data. They can help identify biomarkers and improve diagnosis for various neurological disorders. However, the application of such techniques for rare and heterogeneous diseases remains challenging due to small-cohorts available for acquiring data. Efforts are therefore commonly directed toward improving the classification models, in an effort to optimize outcomes given the limited data. In this study, we systematically evaluated the impact of various machine learning pipeline configurations, including scaling methods, feature selection, dimensionality reduction, and hyperparameter optimization. The efficacy of such components in the pipeline was evaluated on classification performance using multimodal MRI data from a cohort of 16 ALS patients and 14 healthy controls. Our findings reveal that, while certain pipeline components, such as subject-wise feature normalization, help improve classification outcomes, the overall influence of pipeline refinements on performance is modest. Feature selection and dimensionality reduction steps were found to have limited utility, and the choice of hyperparameter optimization strategies produced only marginal gains. Our results suggest that, for small-cohort studies, the emphasis should shift from extensive tuning of these pipelines to addressing data-related limitations, such as progressively expanding cohort size, integrating additional modalities, and maximizing the information extracted from existing datasets. This study provides a methodological framework to guide future research and emphasizes the need for dataset enrichment to improve clinical utility.<\/jats:p>","DOI":"10.3389\/fninf.2025.1568116","type":"journal-article","created":{"date-parts":[[2025,6,13]],"date-time":"2025-06-13T05:49:34Z","timestamp":1749793774000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Evaluating machine learning pipelines for multimodal neuroimaging in small cohorts: an ALS case study"],"prefix":"10.3389","volume":"19","author":[{"given":"Shailesh","family":"Appukuttan","sequence":"first","affiliation":[]},{"given":"Aude-Marie","family":"Grapperon","sequence":"additional","affiliation":[]},{"given":"Mounir Mohamed","family":"El Mendili","sequence":"additional","affiliation":[]},{"given":"Hugo","family":"Dary","sequence":"additional","affiliation":[]},{"given":"Maxime","family":"Guye","sequence":"additional","affiliation":[]},{"given":"Annie","family":"Verschueren","sequence":"additional","affiliation":[]},{"given":"Jean-Philippe","family":"Ranjeva","sequence":"additional","affiliation":[]},{"given":"Shahram","family":"Attarian","sequence":"additional","affiliation":[]},{"given":"Wafaa","family":"Zaaraoui","sequence":"additional","affiliation":[]},{"given":"Matthieu","family":"Gilson","sequence":"additional","affiliation":[]}],"member":"1965","published-online":{"date-parts":[[2025,6,13]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"1620","DOI":"10.1002\/jmri.27019","article-title":"MR biomarkers of degenerative brain disorders derived from diffusion imaging","volume":"52","author":"Andica","year":"2020","journal-title":"J. Magn. Reson. Imaging"},{"key":"B2","doi-asserted-by":"publisher","first-page":"CD004156","DOI":"10.1002\/14651858.CD004156.pub4","article-title":"Treatment for spasticity in amyotrophic lateral sclerosis\/motor neuron disease","volume":"2012","author":"Ashworth","year":"2012","journal-title":"Cochrane Database Syst. Rev"},{"key":"B3","doi-asserted-by":"publisher","first-page":"e0177680","DOI":"10.1371\/journal.pone.0177680","article-title":"Ultra high-field (7tesla) magnetic resonance spectroscopy in amyotrophic lateral sclerosis","volume":"12","author":"Atassi","year":"2017","journal-title":"PLoS ONE"},{"key":"B4","doi-asserted-by":"publisher","first-page":"1","DOI":"10.54294\/uvnhin","article-title":"Advanced normalization tools (ants)","volume":"2","author":"Avants","year":"2009","journal-title":"Insight J"},{"key":"B5","doi-asserted-by":"publisher","first-page":"3289","DOI":"10.1093\/brain\/awad074","article-title":"Defining the phenotypic spectrum of sporadic creutzfeldt-jakob disease mv2k: the kuru plaque type","volume":"146","author":"Baiardi","year":"2023","journal-title":"Brain"},{"key":"B6","doi-asserted-by":"publisher","first-page":"102648","DOI":"10.1016\/j.nicl.2021.102648","article-title":"Ultra-high field (7t) functional magnetic resonance imaging in amyotrophic lateral sclerosis: a pilot study","volume":"30","author":"Barry","year":"2021","journal-title":"Neuroimage Clin"},{"key":"B7","doi-asserted-by":"publisher","first-page":"232","DOI":"10.1080\/21678421.2017.1407795","article-title":"Longitudinal structural changes in als: a three time-point imaging study of white and gray matter degeneration","volume":"19","author":"Bede","year":"2018","journal-title":"Amyotroph. Lateral Scler. Frontotemporal Degener"},{"key":"B8","doi-asserted-by":"publisher","first-page":"293","DOI":"10.1080\/146608200300079536","article-title":"El escorial revisited: revised criteria for the diagnosis of amyotrophic lateral sclerosis","volume":"1","author":"Brooks","year":"2000","journal-title":"Amyotroph. Lateral Scler. Other Motor Neuron Disord"},{"key":"B9","doi-asserted-by":"publisher","first-page":"13","DOI":"10.1016\/S0022-510X(99)00210-5","article-title":"The alsfrs-r: a revised als functional rating scale that incorporates assessments of respiratory function","volume":"169","author":"Cedarbaum","year":"1999","journal-title":"J. Neurol. Sci"},{"key":"B10","doi-asserted-by":"publisher","first-page":"287891","DOI":"10.1155\/2012\/287891","article-title":"Multimodal mri in neurodegenerative disorders","volume":"2012","author":"Cerasa","year":"2011","journal-title":"Neurol. Res. Int"},{"key":"B11","doi-asserted-by":"publisher","first-page":"2879","DOI":"10.1109\/TPAMI.2022.3185773","article-title":"Data augmentation in high dimensional low sample size setting using a geometry-based variational autoencoder","volume":"45","author":"Chadebec","year":"2022","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell"},{"key":"B12","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s41597-024-04157-4","article-title":"Generative models of mri-derived neuroimaging features and associated dataset of 18,000 samples","volume":"11","author":"Chintapalli","year":"2024","journal-title":"Sci. Data"},{"key":"B13","doi-asserted-by":"publisher","first-page":"115","DOI":"10.1016\/j.neuroimage.2019.02.062","article-title":"Benchmarking functional connectome-based predictive models for resting-state fmri","volume":"192","author":"Dadi","year":"2019","journal-title":"NeuroImage"},{"key":"B14","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1016\/j.neuroimage.2010.06.010","article-title":"Automatic parcellation of human cortical gyri and sulci using standard anatomical nomenclature","volume":"53","author":"Destrieux","year":"2010","journal-title":"Neuroimage"},{"key":"B15","doi-asserted-by":"publisher","first-page":"133","DOI":"10.1016\/S0022-510X(01)00610-4","article-title":"The value of muscle exercise in patients with amyotrophic lateral sclerosis","volume":"191","author":"Drory","year":"2001","journal-title":"J. Neurol. Sci"},{"key":"B16","doi-asserted-by":"publisher","first-page":"1395","DOI":"10.1007\/s00234-023-03191-0","article-title":"Association between brain and upper cervical spinal cord atrophy assessed by mri and disease aggressiveness in amyotrophic lateral sclerosis","volume":"65","author":"El Mendili","year":"2023","journal-title":"Neuroradiology"},{"key":"B17","doi-asserted-by":"publisher","first-page":"922","DOI":"10.1136\/adc.65.9.922","article-title":"Magnetic resonance imaging in neurological disorders","volume":"65","author":"Elias-Jones","year":"1990","journal-title":"Arch. Dis. Child"},{"key":"B18","doi-asserted-by":"publisher","first-page":"774","DOI":"10.1016\/j.neuroimage.2012.01.021","article-title":"Freesurfer","volume":"62","author":"Fischl","year":"2012","journal-title":"Neuroimage"},{"key":"B19","doi-asserted-by":"publisher","first-page":"357","DOI":"10.1016\/j.ncl.2010.12.011","article-title":"Advanced MRI in multiple sclerosis: current status and future challenges","volume":"29","author":"Fox","year":"2011","journal-title":"Neurol. Clin"},{"key":"B20","doi-asserted-by":"publisher","first-page":"106391","DOI":"10.1016\/j.compbiomed.2022.106391","article-title":"Data augmentation for medical imaging: a systematic literature review","volume":"152","author":"Garcea","year":"2023","journal-title":"Comput. Biol. Med"},{"key":"B21","doi-asserted-by":"publisher","first-page":"e0316916","DOI":"10.1371\/journal.pone.0316916","article-title":"In vivo mapping of sodium homeostasis disturbances in individual als patients: a brain 23na mri study","volume":"20","author":"Grapperon","year":"2025","journal-title":"PLoS ONE"},{"key":"B22","doi-asserted-by":"publisher","first-page":"422","DOI":"10.1148\/radiol.2019182276","article-title":"Quantitative brain sodium MRI depicts corticospinal impairment in amyotrophic lateral sclerosis","volume":"292","author":"Grapperon","year":"2019","journal-title":"Radiology"},{"key":"B23","doi-asserted-by":"publisher","first-page":"5899","DOI":"10.1093\/cercor\/bhaa165","article-title":"Propagation of bold activity reveals task-dependent directed interactions across human visual cortex","volume":"30","author":"Gravel","year":"2020","journal-title":"Cereb. Cortex"},{"key":"B24","doi-asserted-by":"publisher","first-page":"135","DOI":"10.3389\/fnins.2019.00135","article-title":"Machine learning in amyotrophic lateral sclerosis: achievements, pitfalls, and future directions","volume":"13","author":"Grollemund","year":"2019","journal-title":"Front. Neurosci"},{"key":"B25","doi-asserted-by":"publisher","first-page":"190","DOI":"10.1186\/s44147-024-00522-7","article-title":"Machine learning based parkinson's disease detection and progression evaluation using gray and white matter segmentation from 3D MRI","volume":"71","author":"Islam","year":"2024","journal-title":"J. Eng. Appl. Sci"},{"key":"B26","doi-asserted-by":"publisher","first-page":"614","DOI":"10.1038\/s42003-021-02133-x","article-title":"Integrating molecular, histopathological, neuroimaging and clinical neuroscience data with neuropm-box","volume":"4","author":"Iturria-Medina","year":"2021","journal-title":"Commun. Biol"},{"key":"B27","doi-asserted-by":"publisher","first-page":"81","DOI":"10.3390\/jimaging9040081","article-title":"Deep learning approaches for data augmentation in medical imaging: a review","volume":"9","author":"Kebaili","year":"2023","journal-title":"J. Imaging"},{"key":"B28","doi-asserted-by":"publisher","first-page":"103262","DOI":"10.1016\/j.nicl.2022.103262","article-title":"Advancing brain network models to reconcile functional neuroimaging and clinical research","volume":"36","author":"Kobeleva","year":"2022","journal-title":"Neuroimage Clin"},{"key":"B29","doi-asserted-by":"publisher","first-page":"628","DOI":"10.1136\/jnnp-2015-310998","article-title":"Rate of disease progression: a prognostic biomarker in als","volume":"87","author":"Labra","year":"2016","journal-title":"J. Neurol. Neurosurg. Psychiatry"},{"key":"B30","doi-asserted-by":"publisher","first-page":"1015","DOI":"10.1002\/jmri.27078","article-title":"Artificial intelligence for mr image reconstruction: an overview for clinicians","volume":"53","author":"Lin","year":"2021","journal-title":"J. Magn. Reson. Imaging"},{"key":"B31","doi-asserted-by":"publisher","first-page":"11","DOI":"10.1007\/s13311-016-0484-9","article-title":"Neuroimaging endpoints in amyotrophic lateral sclerosis","volume":"14","author":"Menke","year":"2017","journal-title":"Neurotherapeutics"},{"key":"B32","doi-asserted-by":"publisher","first-page":"1345","DOI":"10.3233\/JAD-220021","article-title":"Machine learning based multimodal neuroimaging genomics dementia score for predicting future conversion to alzheimer's disease","volume":"87","author":"Mirabnahrazam","year":"2022","journal-title":"J. Alzheimers Dis"},{"key":"B33","doi-asserted-by":"publisher","first-page":"576194","DOI":"10.3389\/fneur.2020.576194","article-title":"Comparison of transfer learning and conventional machine learning applied to structural brain mri for the early diagnosis and prognosis of alzheimer's disease","volume":"11","author":"Nanni","year":"2020","journal-title":"Front. Neurol"},{"key":"B34","doi-asserted-by":"publisher","first-page":"106","DOI":"10.1007\/s00330-014-3351-2","article-title":"Brain tumours at 7T MRI compared to 3T\u2014contrast effect after half and full standard contrast agent dose: initial results","volume":"25","author":"Noebauer-Huhmann","year":"2015","journal-title":"Eur. Radiol"},{"key":"B35","doi-asserted-by":"publisher","first-page":"238","DOI":"10.1016\/j.neuroimage.2018.04.070","article-title":"Extracting orthogonal subject-and condition-specific signatures from fmri data using whole-brain effective connectivity","volume":"178","author":"Pallar\u00e9s","year":"2018","journal-title":"NeuroImage"},{"key":"B36","doi-asserted-by":"publisher","first-page":"2456","DOI":"10.1093\/brain\/awr179","article-title":"Sensitivity of revised diagnostic criteria for the behavioural variant of frontotemporal dementia","volume":"134","author":"Rascovsky","year":"2011","journal-title":"Brain"},{"key":"B37","doi-asserted-by":"publisher","first-page":"2849","DOI":"10.1038\/s41598-024-84152-2","article-title":"A comparative machine learning study of schizophrenia biomarkers derived from functional connectivity","volume":"15","author":"Shevchenko","year":"2025","journal-title":"Sci. Rep"},{"key":"B38","doi-asserted-by":"publisher","first-page":"303","DOI":"10.3390\/jimaging8110303","article-title":"Harmonization strategies in multicenter MRI-based radiomics","volume":"8","author":"Stamoulou","year":"2022","journal-title":"J. Imaging"},{"key":"B39","doi-asserted-by":"publisher","first-page":"655","DOI":"10.1007\/978-1-0716-3195-9_22","article-title":"\u201cInterpretability of machine learning methods applied to neuroimaging,\u201d","author":"Thibeau-Sutre","year":"2023","journal-title":"Machine Learning for Brain Disorders, Vol 197"},{"key":"B40","doi-asserted-by":"publisher","first-page":"494","DOI":"10.3174\/ajnr.A8154","article-title":"Neurofilament light chain levels interact with neurodegenerative patterns and motor neuron dysfunction in amyotrophic lateral sclerosis","volume":"45","author":"Tilsley","year":"2024","journal-title":"AJNR Am. J. Neuroradiol"},{"key":"B41","doi-asserted-by":"publisher","first-page":"1310","DOI":"10.1109\/TMI.2010.2046908","article-title":"N4itk: improved n3 bias correction","volume":"29","author":"Tustison","year":"2010","journal-title":"IEEE Trans. Med. Imaging"},{"key":"B42","doi-asserted-by":"publisher","first-page":"38","DOI":"10.1186\/s40035-023-00370-0","article-title":"Multidimensional biomarkers for multiple system atrophy: an update and future directions","volume":"12","author":"Wan","year":"2023","journal-title":"Transl. Neurodegener"},{"key":"B43","doi-asserted-by":"publisher","first-page":"1038","DOI":"10.3389\/fneur.2018.01038","article-title":"Multimodal data and machine learning for detecting specific biomarkers in pediatric epilepsy patients with generalized tonic-clonic seizures","volume":"9","author":"Wang","year":"2018","journal-title":"Front. Neurol"},{"key":"B44","doi-asserted-by":"publisher","first-page":"e083392","DOI":"10.1002\/alz.083392","article-title":"ADNI: Two decades of impact and the path forward","volume":"20","author":"Weiner","year":"2024","journal-title":"Alzheimers Dement"},{"key":"B45","doi-asserted-by":"publisher","first-page":"183","DOI":"10.1016\/j.cpet.2021.09.010","article-title":"Ai-based detection, classification and prediction\/prognosis in medical imaging: towards radiophenomics","volume":"17","author":"Yousefirizi","year":"2022","journal-title":"PET Clin"},{"key":"B46","doi-asserted-by":"publisher","first-page":"947347","DOI":"10.3389\/fneur.2022.947347","article-title":"Structural magnetic resonance imaging findings and histopathological correlations in motor neuron diseases\u2014a systematic review and meta-analysis","volume":"13","author":"Zejlon","year":"2022","journal-title":"Front. Neurol"},{"key":"B47","doi-asserted-by":"publisher","first-page":"421","DOI":"10.3109\/17482968.2011.593036","article-title":"Progression of white matter degeneration in amyotrophic lateral sclerosis: a diffusion tensor imaging study","volume":"12","author":"Zhang","year":"2011","journal-title":"Amyotroph. Lateral Scler"}],"container-title":["Frontiers in Neuroinformatics"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fninf.2025.1568116\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,13]],"date-time":"2025-06-13T05:49:35Z","timestamp":1749793775000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fninf.2025.1568116\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,6,13]]},"references-count":47,"alternative-id":["10.3389\/fninf.2025.1568116"],"URL":"https:\/\/doi.org\/10.3389\/fninf.2025.1568116","relation":{},"ISSN":["1662-5196"],"issn-type":[{"type":"electronic","value":"1662-5196"}],"subject":[],"published":{"date-parts":[[2025,6,13]]},"article-number":"1568116"}}