A machine learning module for the diagnosis of FSH

The muscle involvement encountered in facio-scapulo-humeral myopathy (FSH) is sometimes confusing for the clinician because of its high selectivity and its often asymmetric nature.

Italian researchers have developed an algorithm based on muscle MRI images from 187 patients with FSH and compared to those in a control group:

  • the combination of a precociously affected trapezius muscle and a perfectly preserved subscapularis muscle bilaterally is found to have the best diagnostic value;
  • these results support the use of extended muscle imaging in the upper limbs in the diagnosis of atypical forms of FSH. 

 

Diagnostic magnetic resonance imaging biomarkers for facioscapulohumeral muscular dystrophy identified by machine learning. Monforte M, Bortolani S, Torchia E et al. J Neurol. 2021 (Sept).