Diagnostic accuracy of MRI-derived imaging biomarkers and machine learning-based classification in distal and myofibrillar myopathies
Diagnostische Genauigkeit von aus der MRT abgeleiteten Bildgebungs-Biomarkern und maschinellem Lernen bei distalen und myofibrillären Myopathien
European Radiology · 30.10.2026
DOI: 10.1007/s00330-026-13014-4 ↗Auf einen Blick
Bei 165 Patient:innen zeigten krankheitsspezifische Muskel-MRT-Muster eine hohe diagnostische Genauigkeit. XGBoost erreichte die beste Klassifikationsleistung unter den getesteten Machine-Learning-Modellen.
Fragestellung
Wie genau unterscheiden Muskel-MRT-Muster distale und myofibrilläre Myopathien, und wie gut sind Bildgebungsmerkmale in einem MRT-basierten Diagnosealgorithmus?
Methode
Retrospektive multizentrische Analyse von Muskel-MRT und CT bei genetisch definierten distalen und myofibrillären Myopathien; Musterbeschreibung, Berechnung von Sensitivität und Spezifität sowie Training überwachter Machine-Learning-Algorithmen.
Ergebnisse
165 Patient:innen wurden analysiert, darunter 138 molekular bestätigte Fälle mit 19 Krankheitsgenen. Alle Bildgebungsmuster hatten eine Sensitivität und Spezifität von 75 bis 100 %. XGBoost unterschied 11 genetische Entitäten mit 69 % balancierter Genauigkeit.
Schlussfolgerung der Autor:innen
Distale und myofibrilläre Myopathien zeigen reproduzierbare, krankheitsspezifische Bildgebungssignaturen mit hoher diagnostischer Genauigkeit. Die Integration in Machine-Learning-Modelle kann die Priorisierung genetischer Diagnosen erleichtern.
Kernaussagen
- Muskel-MRT-Muster waren krankheitsspezifisch und diagnostisch aussagekräftig.
- Sensitivität und Spezifität lagen für alle Muster bei 75 bis 100 %.
- XGBoost erreichte 69 % balancierte Genauigkeit bei 11 genetischen Entitäten.
Deutsche Zusammenfassung: KI-gestützt aus dem Original-Abstract erstellt; alle Zahlen wurden automatisch mit dem Original abgeglichen. Maßgeblich ist der Original-Abstract.
Kurz gefasst
Fachartikel in der Zeitschrift European Radiology, online erschienen am 30.10.2026. Verfasst von Sara Bortolani und 35 weiteren Autor:innen. Beteiligte Einrichtung: UOC di Neurologia, Fondazione Policlinico Universitario "A. Gemelli" IRCCS, Rome, Italy. Die Arbeit nennt 39 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: AFM-Téléthon, Academy of Medical Sciences Professorship Scheme, NIHR Newcastle Biomedical Research Centre.
Abstract (Originalsprache)
Objectives
To characterise and validate disease-specific muscle imaging patterns in distal and myofibrillar myopathies, quantify their diagnostic performance, and develop an imaging-based classification framework incorporating machine learning. Materials and methods Muscle MRI and CT scans from a multicentre national cohort of patients with genetically defined distal and myofibrillar myopathies were retrospectively analysed. The pattern of muscle involvement for each disease entity was described, and its sensitivity and specificity calculated. Supervised machine learning algorithms were trained using muscle scores and imaging-derived features, and their diagnostic performance was compared using balanced accuracy.
Results
A total of 165 patients (mean age 48 ± 16 years; 87 men), including 138 molecularly confirmed cases spanning 19 disease genes, were analysed. Overall, 6350 muscles were systematically evaluated. Key findings included early biceps femoris short head involvement in GNE myopathy, “fat pockets” in myotilinopathy, and early flexor hallucis longus involvement in desminopathy. All imaging patterns showed high sensitivity and specificity (from 75 to 100%). Among the tested machine learning models, XGBoost achieved the highest performance, correctly distinguishing among 11 genetic disease entities with a balanced accuracy of 69%.
Conclusion
Distal and myofibrillar myopathies demonstrate reproducible disease-specific imaging signatures with high diagnostic accuracy. Integration of these imaging biomarkers into a machine learning framework enables systematic prioritisation of likely genetic diagnoses and may facilitate diagnostic work-up in these heterogeneous inherited muscle disorders. Key Points Question How accurately do muscle MRI patterns distinguish distal and myofibrillar myopathies, and how do imaging features perform in an updated MRI-based diagnostic algorithm ?
Findings
Disease-specific muscle MRI patterns showed high diagnostic accuracy, while XGBoost achieved the highest balanced accuracy among machine learning models for imaging-based disease classification . Clinical relevance Refined muscle MRI patterns provide diagnostically useful imaging biomarkers in distal and myofibrillar myopathies. Their integration into machine learning-based classification models may facilitate precision diagnostics and guide targeted genetic testing .
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- UOC di Neurologia, Fondazione Policlinico Universitario "A. Gemelli" IRCCS, Rome, Italy.
- Università Cattolica del Sacro Cuore, Rome, Italy.
- John Walton Muscular Dystrophy Research Centre, NIHR Biomedical Research Centre: Newcastle, Newcastle University and Newcastle Hospitals NHS Foundation Trust, Newcastle upon Tyne, UK.
- Neuroimmunology and Neuromuscular Disorders Unit, Fondazione IRCCS Istituto Neurologico Carlo Besta, Milan, Italy.
- Neuroradiology, Fondazione IRCCS Istituto Neurologico Carlo Besta, Milan, Italy.
- Department of Neurosciences, Neuromuscular Center, University of Padova, Padova, Italy.
Förderung
- AFM-Téléthon
- Academy of Medical Sciences Professorship Scheme
- NIHR Newcastle Biomedical Research Centre
Zitieren
APA
Bortolani, S., Monforte, M., Verdu-Diaz, J., Bonanno, S., Moscatelli, M., Riguzzi, P., Lupi, A., Torchia, E., Gadaleta, G., Vattemi, G., Ravaglia, S., Valentino, M. L., Falzone, Y., Primiano, G., Lopergolo, D., Lucchini, M., Petrucci, A., Torri, F., Garibaldi, M., …, & Tasca, G. (2026). Diagnostic accuracy of MRI-derived imaging biomarkers and machine learning-based classification in distal and myofibrillar myopathies. European Radiology. https://doi.org/10.1007/s00330-026-13014-4
Vancouver
Bortolani S, Monforte M, Verdu-Diaz J, Bonanno S, Moscatelli M, Riguzzi P, et al. Diagnostic accuracy of MRI-derived imaging biomarkers and machine learning-based classification in distal and myofibrillar myopathies. European Radiology. 2026. doi:10.1007/s00330-026-13014-4
BibTeX
@article{Bortolani2026,
author = {Bortolani, Sara and Monforte, Mauro and Verdu-Diaz, Jose and Bonanno, Silvia and Moscatelli, Marco and Riguzzi, Pietro and Lupi, Amalia and Torchia, Eleonora and Gadaleta, Giulio and Vattemi, Gaetano and Ravaglia, Sabrina and Valentino, Maria Lucia and Falzone, Yuri and Primiano, Guido and Lopergolo, Diego and Lucchini, Matteo and Petrucci, Antonio and Torri, Francesca and Garibaldi, Matteo and Ravera, Beatrice and Siciliano, Gabriele and Mirabella, Massimiliano and De Stefano, Nicola and Servidei, Serenella and Rodolico, Carmelo and Gerevini, Simonetta and Previtali, Stefano C and Liguori, Rocco and Tonin, Paola and Mongini, Tiziana E and Pegoraro, Elena and Diaz-Manera, Jordi and Savarese, Marco and Maggi, Lorenzo and Ricci, Enzo and Tasca, Giorgio},
title = {Diagnostic accuracy of MRI-derived imaging biomarkers and machine learning-based classification in distal and myofibrillar myopathies},
journal = {European Radiology},
year = {2026},
doi = {10.1007/s00330-026-13014-4}
}
EUVIMED