BACKGROUND: Neuromuscular diseases (NMDs) are rare disorders characterized by progressive muscle fibre loss, leading to replacement by fibrotic and fatty tissue, muscle weakness and disability. Early diagnosis is critical for therapeutic decisions, care planning and genetic counselling. Muscle magnetic resonance imaging (MRI) has emerged as a valuable diagnostic tool by identifying characteristic patterns of muscle involvement. However, the increasing complexity of these patterns complicates their interpretation, limiting their clinical utility. Additionally, multi-study data aggregation introduces heterogeneity challenges. This study presents a novel multi-study harmonization pipeline for muscle MRI and an AI-driven diagnostic tool to assist clinicians in identifying disease-specific muscle involvement patterns. METHODS: We developed a preprocessing pipeline to standardize MRI fat content across datasets, minimizing source bias. An ensemble of XGBoost models was trained to classify patients based on intramuscular fat replacement, age at MRI and sex. The SHapley Additive exPlanations (SHAP) framework was adapted to analyse model predictions and identify disease-specific muscle involvement patterns. To address class imbalance, training and evaluation were conducted using class-balanced metrics. The model's performance was compared against four expert clinicians using 14 previously unseen MRI scans. RESULTS: Using our harmonization approach, we curated a dataset of 2961 MRI samples from genetically confirmed cases of 20 paediatric and adult NMDs. The model achieved a balanced accuracy of 64.8% ± 3.4%, with a weighted top-3 accuracy of 84.7% ± 1.8% and top-5 accuracy of 90.2% ± 2.4%. It also identified key features relevant for differential diagnosis, aiding clinical decision-making. Compared to four expert clinicians, the model obtained the highest top-3 accuracy (75.0% ± 4.8%). The diagnostic tool has been implemented as a free web platform, providing global access to the medical community. CONCLUSIONS: The application of AI in muscle MRI for NMD diagnosis remains underexplored due to data scarcity. This study introduces a framework for dataset harmonization, enabling advanced computational techniques. Our findings demonstrate the potential of AI-based approaches to enhance differential diagnosis by identifying disease-specific muscle involvement patterns. The developed tool surpasses expert performance in diagnostic ranking and is accessible to clinicians worldwide via the Myo-Guide online platform.

Advanced Imaging and AI Center Mondino IRCCS Foundation Pavia Italy

Aix Marseille University CRMBM CNRS UMR 7339 Marseille France

Biomedical Research Institute Sant Pau Barcelona Spain

C J Gorter MRI Center Department of Radiology Leiden University Medical Center Leiden The Netherlands

Centre de Référence des Maladies du Motoneurone Department of Neurology Montpellier University Hospital Montpellier France

Centro de Investigaciones Biomédicas en Red en Enfermedades Raras Madrid Spain

Copenhagen Neuromuscular Centre Rigshospitalet Copenhagen University Hospital Copenhagen Denmark

Department of Brain and Behavioural Sciences University of Pavia Pavia Italy

Department of Genetics Children's Hospital of Eastern Ontario Ottawa Canada

Department of Medicine The Ottawa Hospital Ottawa Canada

Department of Neurology Faculdade de Medicina da Universidade de São Paulo São Paulo Brazil

Department of Neurology Huashan Hospital Fudan University Shanghai China

Department of Neurology Leiden University Medical Center Leiden The Netherlands

Department of Neurology Pusan National University School of Medicine Busan Republic of Korea

Department of Neuroradiology Great Ormond Street Hospital for Children NHS Foundation Trust London UK

Department of Neuroradiology I2FH Platform Montpellier University Hospital Montpellier France

Department of Neuroscience Mental Health and Sensory Organs SAPIENZA University of Rome Rome Italy

Department of Translational Research and of New Surgical and Medical Technologies University of Pisa Pisa Italy

Dubowitz Neuromuscular Centre UCL Great Ormond Street Institute of Child Health and Great Ormond Street Hospital London UK

Fondazione Policlinico Universitario Agostino Gemelli Rome Italy

Hospital Clínico Universidad de Chile Santiago de Chile Chile

Hospital Universitari Vall d'Hebron Barcelona Spain

Interdisciplinary Computing and Complex BioSystems Research Group School of Computing Newcastle University Newcastle upon Tyne UK

John Walton Muscular Dystrophy Research Centre Newcastle University Newcastle upon Tyne UK

Leeds Teaching Hospitals NHS Trust Leeds UK

National Institute of Mental Health and Neurosciences Bengaluru India

Neurology Department Shariati Hospital Neuromuscular Research Center Tehran University of Medical Sciences Tehran Iran

Neuromuscular and Rare Disease Centre Neurology Unit Sant'Andrea Hospital Rome Italy

Neuromuscular Disease Unit Neurology Department Hospital Universitario Nuestra Señora de Candelaria Tenerife Spain

Neuromuscular Disorders Unit Department of Neurology Hospital de la Santa Creu i Sant Pau Barcelona Spain

Neuromuscular Disorders Unit Neurology Department Hospital 12 de Octubre Madrid Spain

Neuromuscular Disorders Unit Neurology Department Hospital Universitari Vall d'Hebron Barcelona Spain

Neuromuscular Reference Center Department of Neurology Universitair Ziekenhuis van Antwerpen Universiteit Antwerpen Antwerp Belgium

Northern Care Alliance NHS Foundation Trust Manchester UK

Ottawa Hospital Research Institute Ottawa Canada

Paris Est University APHP Henri Mondor University Hospital Créteil France

Pediatric Neurology Department of Woman and Child Health and Public Health Child Health Area Università Cattolica del Sacro Cuore Rome Italy

Programa de Doctorado en Ciencias Médicas y Especialidad Escuela de Postgrado Facultad de Medicina Universidad de Chile Santiago Chile

Reference Center for Neuromuscular Disorders CHU La Timone Aix Marseille University Marseille France

St George's University and St George's University Hospitals NHS Foundation Trust London UK

Translational and Clinical Research Institute Newcastle University Newcastle upon Tyne UK

University Hospital Brno Brno Czech Republic

University Hospital Raymond Poincaré Garches France

University of Pavia; Mondino IRCCS Foundation Pavia Italy

UOC di Neurologia Fondazione Policlinico Universitario Agostino Gemelli IRCCS Rome Italy

Zobrazit více v PubMed

Mercuri E. and Muntoni F., “Muscular Dystrophies,” Lancet 381, no. 9869 (2013): 845–860. PubMed

Nuñez‐Peralta C., Alonso‐Pérez J., and Díaz‐Manera J., “The Increasing Role of Muscle MRI to Monitor Changes Over Time in Untreated and Treated Muscle Diseases,” Current Opinion in Neurology 33, no. 5 (2020): 611–620. PubMed

Dahlqvist J. R., Widholm P., Leinhard O. D., and Vissing J., “MRI in Neuromuscular Diseases: An Emerging Diagnostic Tool and Biomarker for Prognosis and Efficacy,” Annals of Neurology 88, no. 4 (2020): 669–681. PubMed

Burakiewicz J., Sinclair C. D. J., Fischer D., Walter G. A., Kan H. E., and Hollingsworth K. G., “Quantifying Fat Replacement of Muscle by Quantitative MRI in Muscular Dystrophy,” Journal of Neurology 264, no. 10 (2017): 2053–2067. PubMed PMC

Pezeshk P., Alian A., and Chhabra A., “Role of Chemical Shift and Dixon Based Techniques in Musculoskeletal MR Imaging,” European Journal of Radiology 94 (2017): 93–100. PubMed

Mercuri E., Talim B., Moghadaszadeh B., et al., “Clinical and Imaging Findings in Six Cases of Congenital Muscular Dystrophy With Rigid Spine Syndrome Linked to Chromosome 1p (RSMD1),” Neuromuscular Disorders 12, no. 7 (2002): 631–638. PubMed

Tasca G., Monforte M., Díaz‐Manera J., et al., “MRI in Sarcoglycanopathies: A Large International Cohort Study,” Journal of Neurology, Neurosurgery, and Psychiatry 89, no. 1 (2018): 72–77. PubMed

Bugiardini E., Morrow J. M., Shah S., et al., “The Diagnostic Value of MRI Pattern Recognition in Distal Myopathies,” Frontiers in Neurology 9 (2018): 456. PubMed PMC

Díaz‐Manera J., Llauger J., Gallardo E., and Illa I., “Muscle MRI in Muscular Dystrophies,” Acta Myologica 34, no. 2–3 (2015): 95. PubMed PMC

Carlier P. G. and Reyngoudt H., “The Expanding Role of MRI in Neuromuscular Disorders,” Nature Reviews. Neurology 16, no. 6 (2020): 301–302. PubMed

Wei P., Zhong H., Xie Q., et al., “Machine Learning‐Based Radiomics to Differentiate Immune‐Mediated Necrotizing Myopathy From Limb‐Girdle Muscular Dystrophy R2 Using MRI,” Frontiers in Neurology 14 (2023): 1251025. PubMed PMC

Nagawa K., Suzuki M., Yamamoto Y., et al., “Texture Analysis of Muscle MRI: Machine Learning‐Based Classifications in Idiopathic Inflammatory Myopathies,” Scientific Reports 11, no. 1 (2021): 9821. PubMed PMC

Monforte M., Bortolani S., Torchia E., et al., “Diagnostic Magnetic Resonance Imaging Biomarkers for Facioscapulohumeral Muscular Dystrophy Identified by Machine Learning,” Journal of Neurology 269, no. 4 (2022): 2055–2063. PubMed

Bolano‐Diaz C., Verdú‐Díaz J., Gonzalez‐Chamorro A., et al., “Magnetic Resonance Imaging‐Based Criteria to Differentiate Dysferlinopathy From Other Genetic Muscle Diseases,” Neuromuscular Disorders 34 (2024): 54–60. PubMed

Verdú‐Díaz J., Alonso‐Pérez J., Nuñez‐Peralta C., et al., “Accuracy of a Machine Learning Muscle MRI‐Based Tool for the Diagnosis of Muscular Dystrophies,” Neurology 94, no. 10 (2020): e1094–e1102. PubMed

Esteller D., Schiava M., Verdú‐Díaz J., et al., “Analysis of Muscle Magnetic Resonance Imaging of a Large Cohort of Patient With VCP‐Mediated Disease Reveals Characteristic Features Useful for Diagnosis,” Journal of Neurology 270, no. 12 (2023): 5849–5865. PubMed PMC

Warman Chardon J., Díaz‐Manera J., Tasca G., et al., “MYO‐MRI Diagnostic Protocols in Genetic Myopathies,” Neuromuscular Disorders 29, no. 11 (2019): 827–841. PubMed

Breiman L., “Random Forests,” Machine Learning 45, no. 1 (2001): 5–32.

Laaksonen J. and Oja E., “Classification With Learning k‐Nearest Neighbors,” in Proceedings of International Conference on Neural Networks (ICNN'96), (1996): 1480–1483.

Jacobsen L. N., Stemmerik M. G., Skriver S. V., Pedersen J. J., Løkken N., and Vissing J., “Contractile Properties and Magnetic Resonance Imaging‐Assessed fat Replacement of Muscles in Myotonia Congenita,” European Journal of Neurology 31, no. 4 (2024): e16207. PubMed PMC

Branco P., Torgo L., and Ribeiro R. P., “A Survey of Predictive Modeling on Imbalanced Domains,” ACM Computing Surveys 49, no. 2 (2016): 1–50.

Lundberg S. M., Erion G., Chen H., et al., “From Local Explanations to Global Understanding With Explainable AI for Trees,” Nature Machine Intelligence 2, no. 1 (2020): 56–67. PubMed PMC

Liu C.‐Y., Yao J., Kovacs W. C., et al., “Skeletal Muscle Magnetic Resonance Biomarkers in GNE Myopathy,” Neurology 96, no. 5 (2021): e798–e808. PubMed PMC

Tasca G., Ricci E., Monforte M., et al., “Muscle Imaging Findings in GNE Myopathy,” Journal of Neurology 259, no. 7 (2012): 1358–1365. PubMed

Angelini C., Fanin M., Freda M. P., Duggan D. J., Siciliano G., and Hoffman E. P., “The Clinical Spectrum of Sarcoglycanopathies,” Neurology 52, no. 1 (1999): 176. PubMed

Widholm P., Ahlgren A., Karlsson M., et al., “Quantitative Muscle Analysis in Facioscapulohumeral Muscular Dystrophy Using Whole‐Body Fat‐Referenced MRI: Protocol Development, Multicenter Feasibility, and Repeatability,” Muscle & Nerve 66, no. 2 (2022): 183–192. PubMed

Engelke K., Chaudry O., Gast L., et al., “Magnetic Resonance Imaging Techniques for the Quantitative Analysis of Skeletal Muscle: State of the Art,” Journal of Orthopaedic Translation 42 (2023): 57–72. PubMed PMC

Agosti A., Shaqiri E., Paoletti M., et al., “Deep Learning for Automatic Segmentation of Thigh and Leg Muscles,” Magma 35, no. 3 (2022): 467–483. PubMed PMC

Felisaz P. F., Colelli G., Ballante E., et al., “Texture Analysis and Machine Learning to Predict Water T2 and Fat Fraction From Non‐Quantitative MRI of Thigh Muscles in Facioscapulohumeral Muscular Dystrophy,” European Journal of Radiology 134 (2021): 109460. PubMed

Trueb P., Getzmann J. M., Ried E., Deininger‐Czermak E., Garcia Schueler H. I., and Guggenberger R., “Comparison of Muscle fat Fraction Measurements in the Lower Spine Musculature With Non‐Contrast‐Enhanced CT and Different MR Imaging Sequences,” European Journal of Radiology 150 (2022): 110260. PubMed

Sarkozy A., Deschauer M., Carlier R. Y., et al., “Muscle MRI Findings in Limb Girdle Muscular Dystrophy Type 2L,” Neuromuscular Disorders 22 (2012): S122–S129. PubMed

Ropars J., Gravot F., Ben Salem D., Rousseau F., Brochard S., and Pons C., “Muscle MRI: A Biomarker of Disease Severity in Duchenne Muscular Dystrophy? A Systematic Review,” Neurology 94, no. 3 (2020): 117–133. PubMed

Nicolau S. and Naddaf E., “Muscle MRI for Neuromuscular Disorders Using Muscle MRI to Diagnose Neuromuscular Conditions Requires Awareness of Different Patterns of Muscle Involvement,” (2020).

Vivekanandam V., Suetterlin K., Matthews E., et al., “Muscle MRI in Periodic Paralysis Shows Myopathy Is Common and Correlates With Intramuscular Fat Accumulation,” Muscle & Nerve 68, no. 4 (2023): 439–450. PubMed

Finlayson S., Morrow J. M., Rodriguez Cruz P. M., et al., “Muscle Magnetic Resonance Imaging in Congenital Myasthenic Syndromes,” Muscle & Nerve 54, no. 2 (2016): 211–219. PubMed PMC

Tasca G., Monforte M., De Fino C., Kley R. A., Ricci E., and Mirabella M., “Magnetic Resonance Imaging Pattern Recognition in Sporadic Inclusion‐Body Myositis,” Muscle & Nerve 52, no. 6 (2015): 956–962. PubMed

Mercuri E., Lampe A., Allsop J., et al., “Muscle MRI in Ullrich Congenital Muscular Dystrophy and Bethlem Myopathy,” Neuromuscular Disorders 15, no. 4 (2005): 303–310. PubMed

Tarazona S., Arzalluz‐Luque A., and Conesa A., “Undisclosed, Unmet and Neglected Challenges in Multi‐Omics Studies,” Nature Computational Science 1, no. 6 (2021): 395–402. PubMed

Yu G., Li Q., Shen D., and Liu Y., “Optimal Sparse Linear Prediction for Block‐Missing Multi‐Modality Data Without Imputation,” Journal of the American Statistical Association 115, no. 531 (2020): 1406–1419. PubMed PMC

Zhang Y., Tang N., and Qu A., “Imputed Factor Regression for High‐Dimensional Block‐Wise Missing Data,” Statistica Sinica 30, no. 2 (2020): 631–651.