Download PDF
Semin Musculoskelet Radiol 2025; 29(05): 671-681
DOI: 10.1055/s-0045-1809941
Review Article

How Can Artificial Intelligence Help Avoid Mistakes in Musculoskeletal Imaging?

Authors

  • Marie Pauline Talabard 

    1   Cochin Hospital, Paris, France
    5   Department of Musculoskeletal Radiology, Hôpital Cochin, AP-HP Centre, Université Paris Cité, Paris, France

  • Nor-Eddine Regnard 

    2   Réseau d'Imagerie Sud Francilien, Lieusaint, France Gleamer, Paris, France

  • Patrick Omoumi 

    3   Department of Radiology, Lausanne University Hospital and University of Lausanne, Lausanne, Switzerland

  • Pedro Augusto Gondim Texeira

    4   Guilloz Imaging Department, Central Hospital, University Hospital Center of Nancy, Nancy, France

  • Antoine Feydy 

    5   Department of Musculoskeletal Radiology, Hôpital Cochin, AP-HP Centre, Université Paris Cité, Paris, France
Further Information
Also available at
   Permissions and Reprints
Preview

Abstract

Musculoskeletal imaging plays a central role in diagnosing and managing a wide range of orthopedic conditions. However, it remains susceptible to both interpretive and noninterpretive errors, amplified by increasing imaging demand and complexity. Artificial intelligence, especially deep learning and large language models, has shown growing potential to reduce these errors at every stage of the imaging workflow. From optimizing exam requests and imaging protocols to reducing artifacts and improving interpretative consistency, artificial intelligence supports radiologists in enhancing diagnostic accuracy, efficiency, and reproducibility. Applications now extend across all modalities, including magnetic resonance, radiography, computed tomography, and ultrasound, and they address common pitfalls such as subjective assessments and measurement variability. Post-interpretation tools using large language models further improve report clarity and patient communication. Although integration into clinical practice remains ongoing, artificial intelligence already offers a transformative opportunity to improve musculoskeletal imaging quality and safety through collaborative human–machine interaction.

Keywords

musculoskeletal - artificial intelligence - large language model - convolutional neural network - machine learning


Publication History

Article published online:
07 October 2025

© 2025. Thieme. All rights reserved.

Thieme Medical Publishers, Inc.
333 Seventh Avenue, 18th Floor, New York, NY 10001, USA