Machine learning outperforms human experts in MRI pattern analysis of muscular dystrophies

Jasper M. Morrow, Maria Pia Sormani

Research output: Contribution to journalEditorialpeer-review

Abstract

Accurate genetic diagnosis in inherited muscle diseases is challenging, with over 30 causative genes for the limb-girdle muscular dystrophies (LGMDs) alone,1 but also vital, with gene-specific therapies now available and more on the horizon.2 Although next-generation sequencing provides a wealth of genetic data, interpretation often requires deep phenotyping including clinical, laboratory, and imaging data. Many MRI studies have described the different patterns of muscle involvement in inherited muscle diseases.3 Whereas the MRI patterns are generally concordant with the clinical distribution of weakness, using MRI patterns of involvement potentially offers greater complexity and discrimination because of the ability to assess each muscle individually.

Original languageEnglish
Pages (from-to)421-422
Number of pages2
JournalNeurology
Volume94
Issue number10
DOIs
Publication statusPublished - Mar 10 2020

ASJC Scopus subject areas

  • Clinical Neurology

Cite this