O04 Myoguide.org: a web-based portal supporting the analysis of MRIs for the diagnosis of neuromuscular patients
Muscle magnetic resonance imaging (MRI) is a useful tool to study muscle structure in patients with genetic neuromuscular diseases (NMDs), as it can easily identify fat replacement and changes in water content. MRI has been extensively used for the diagnosis of NMDs. However, the analysis of MRIs an...
Saved in:
| Published in: | Neuromuscular disorders : NMD Vol. 33; p. S68 |
|---|---|
| Main Authors: | , , , , , |
| Format: | Journal Article |
| Language: | English |
| Published: |
Elsevier B.V
01-10-2023
|
| Online Access: | Get full text |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Muscle magnetic resonance imaging (MRI) is a useful tool to study muscle structure in patients with genetic neuromuscular diseases (NMDs), as it can easily identify fat replacement and changes in water content. MRI has been extensively used for the diagnosis of NMDs. However, the analysis of MRIs and the identification of specific patterns of muscle involvement can be a complicated task that requires a high degree of specialization.
Our aim was to create a web platform to help in the diagnosis of NMDs using MRI. In order to do this, we included MYO-Guide, a machine learning based algorithm that predicts the diagnosis of patients with NMDs using MRIs, but also an information catalogue including examples of typical patterns of muscle involvement as well as reports describing how to analyse MRIs. The web www.myoguide.org has been launched thanks to the collaboration of 30 centres worldwide. The web contains MYO-Guide algorithm which is able to predict the diagnosis of 15 different neuromuscular diseases with an accuracy of 85%. We have collected 2307 muscle MRIs from the pelvis and/or lower limbs of different muscle diseases that we have used to build the algorithm. Moreover, the web contains information about the pattern of involvement of different diseases at different stages of disease progression as well as a large number of articles supporting training of health care professionals interested in the analysis of MRIs. We are now working on the development of an automatic segmentation tool that should be able to identify individual muscles and automatically quantify fat content. Myoguide.org is a webpage that contains several resources to help in the analysis of muscle MRIs, including an artificial intelligence predicting tool and many training resources for the medical community. |
|---|---|
| ISSN: | 0960-8966 1873-2364 |
| DOI: | 10.1016/j.nmd.2023.07.019 |