About

What is Myo-Guide?

MYO-Guide is a platform that aims to push the role of Magnetic Ressonance Imaging (MRI) in the diagnosis and study of Neuromuscular Diseases (NMDs). To learn more about the role of MRI in NMDs, please visit the following link. The main tool offered by MYO-Guide is an online Machine Learning model for diagnosing NMDs using the Mercuri score. For a full list of MYO-Guide features, please visit the following link.

MYO-Guide started in 2018 as a study of the efficacy of machine learning in the diagnostic process of neuromuscular diseases. In that project, the Mercuri scores of the pelvis, thighs, and lower legs from more than 1000 patients were collected. These patients had a confirmed diagnosis ranging between 10 different NMDs:

• Duchenne and Becker muscular dystrophy
• Limb girdle muscular dystrophy R1 or calpainopathy
• Limb girdle muscular dystrophy R2 or dysferlinopathy
• Limb girdle muscular dystrophy R3 to R6 or sarcoglycanopathies
• Limb girdle muscular dystrophy R9 or fukutin-related disease
• Limb girdle muscular dystrophy R12 or anoctaminopathy
• Limb girdle muscular dystrophy produced by mutations in LMNA gene
• Oculopharyngeal muscular dystrophy
• Facio-scapulo humeral muscular dystrophy (FSHD)
• Pompe disease

A Random Forest model was trained, and obtained an accuracy of 95.7%. This first project was described in the following paper:

Article

Accuracy of a machine learning muscle MRI-based tool for the diagnosis of muscular dystrophies

José Verdú-Díaz, Jorge Alonso-Pérez, Claudia Nuñez-Peralta, Giorgio Tasca, John Vissing, Volker Straub, Roberto Fernández-Torrón, Jaume Llauger, Isabel Illa, Jordi Díaz-Manera

First published February 6, 2020, DOI: https://doi.org/10.1212/WNL.0000000000009068

Now, we are working towards improving the first version of the model. Our main aim is to increase the number of diseases that Myo-Guide can predict from 10 to at least 25 including among others myotonic dystrophy type I, FSHD type II, spinal muscular atrophy (SMA), myofibrillar myopathies or necrotizing myopathy:

• Dystrophynopathy
• CAPN3 (LGMD-R1)
• DYSF (LGMD-R2)
• SGCA (LGMD-R3)
• SGCB (LGMD-R4)
• SGCG (LGMD-R5)
• TTN
• ANO5 (LGMD-R12)
• FKRP (LGMD-R9)
• COL6A1, A2 and A3 (Bethlem and Ullrich)
• MYOT (Myofibrillar myopathy type I)
• RYR-1
• DMPK (Myotonic dystrophy type I)
• ZNF9 (Myotonic dystrophy type II)
• SMN-1 (Spinal muscle atrophy)
• MYH7
• GNE (Nonaka disease)
• GAA (Pompe disease)
• LMNA
• PABP2 (Oculopharyngeal muscle dystrophy)
• D4Z4 deletion (Facio-Scapulo-Humeral muscular dystrophy type 1)
• SMCHD1 (Facio-Scapulo-Humeral muscular dystrophy type 2)
• VCP (HIBM + Paget)
• Necrotizing myopathy (SRP/HMGCoA)
• IBM (Inclusion Body Myositis)

The diagnosis tool of Myo-Guide is available online on the following link, and will allow doctors from everywhere in the world to use Myo-Guide without requiring any software. The tool attempts to be as user-friendly as possible, extensive documentation on how to use it is available on the following link. As we are conscious of how difficult it is to find enough time to calculate Mercuri score for each muscle, we also want to provide users with an automatic segmentation tool that can automatically identify the muscles and calculate the Mercuri score.

ISRCTN Registry

MYO-Guide is registered in the ISRCTN Registry with the following DOI: doi.org/10.1186/ISRCTN14323809.

The MYO-Guide Consortium

In order to achieve our goal we are seeking for collaboration with clinicians worldwide that have muscle MRIs of patients with a confirmed diagnosis of any of the listed disorders. To collect the MRIs we use MYO-Share, an online platform for sharing MRIs between researchers built on the Azure platform that esures the security and integrity of the images uploaded. In case you are interested collaborating, please contact us at: [email protected]

MYO-Guide wouldn't be possible without our international collaborator network. We owe a big thanks to all the centres, clinicians and researchers that have kindly shared their data with us.

The consortium is currently formed by:

• Hospital Italiano de Buenos Aires (Buenos Aires, Argentina)
• Dr. Marcelo Rugiero
• Mariela Bettini
• Ottawa NeuroMuscular Centre (Ottawa, Canada)
• Dr. Jodi Warman
• Dr. Ian Smith
• Hospital Clínico Universidad de Chile y Clínica Dávila (Santiago de Chile, Chile)
• Dr. Jorge Bevilacqua Rivas
• Dr. Jorge Díaz-Jara
• Dr. Nicholas Earle
• Gabriel Aguilera
• Benjamin Pizzarro
• Huashan Hospital Fudan University (Shangai, China)
• Dr. Sushan Luo
• Huahua Zhong
• University Hospital Brno (Brno, Czech Republic)
• Dr. Peter Krkoska
• Righospitalet University of Copenhagen (Copenhagen, Denmark)
• Prof. John Vissing
• Dr. Nanna Scharf Poulsen
• Fondazione Policlinico Universitario Agostino Gemelli (Rome, Italy)
• Dr. Mauro Monforte
• Prof. Giorgio Tasca
• Prof. Enzo Ricci
• Sapienza Sant'Andrea University Hospital (Rome, Italy)
• Dr. Matteo Garibaldi
• Dr. Laura Fionda
• Dr. Rocco Constanzo
• Fondazione Istituto Neurologico Casimiro Mondino (Pavia, Italy)
• Dr. Anna Pichiecchio
• Silvia Nicolosi
• IRCCS Fondazione Stella Maris Universita Di Pisa (Pisa, Italy)
• Dr. Bianca Buchignani
• Dr. Guja Astrea
• Prof. Roberta Battini
• Dr. Rosa Pasquariello
• Radboud University Medical Center (Nijmegen, Netherlands)
• Dr. Nicol Voermans
• Leiden University Medical Center (Leiden, Netherlands)
• Dr. Hermien Kan
• Dr. Donnie Cameron
• Kyungpook National University Chilgok Hospital (Daegu, South Korea)
• Dr. Jin-Sung Park
• Dr. Jong-Mok Lee
• Yangsan University Hospital Pusan National University (Pusan, South Korea)
• Dr. Jin-Hong Shin
• Dr. Young-Eun Park
• Hospital Clinic de Barcelona (Barcelona, Spain)
• Dr. Diana Esteller
• Fundacion Investigacion Biomedica Hospital 12 de Octubre (Madrid, Spain)
• Dr. Cristina Domínguez González
• Laura Bermejo Guerrero
• Great Ormond Street Hospital (London, UK)
• Dr. Anna Sarkozy
• Dr. Giovanni Baranello
• Jain-COS Study Newcastle University (Newcastle upon Tyne, UK)
• Prof. Volker Straub
• Dr. Ian Wilson
• Newcastle Magnetic Resonance Centre Newcastle University (Newcastle upon Tyne, UK)
• Dr. Kieren Hollingsworth
• The Newcastle upon Tyne Hospitals (Newcastle upon Tyne, UK)
• Prof. Jordi Diaz Manera
• Northern Care Alliance NHS Trust (Manchester, UK)
• Prof. Mark Roberts
• Dr. James Lilleker
• University College London Hospitals (London, UK)
• Dr. Jasper Morrow
• Leeds Teacing Hospitals (Leeds, UK)
• Dr. Anne Marie Childs
• St George's University Hospitals (London, UK)
• Dr. Emma Matthews
• Shariati Hospital (Tehran, Iran)
• Prof. Shahriar Nafissi
• Ali Okhovat
• NIMHANS (Bengaluru, India)
• Dr. Atchayaram nalini
• Edamkanti Sravan Kumar Reddy
• Ms Dipti Baskar
• Ms Saraswati Nashi
• Antwerp University Hospital (Antwerp, Belgium)
• Alicia Alonso
• Hospital Universitario Nuestra Señora de Candelaria (Tenerife, Spain)
• Jorge Alonso-Pérez

What does MYO-Guide offer?

MYO-Guide acts as an umbrella for different projects involving MRI and NMDs:

• Diagnosis Tool: online tool for diagnosing NMDs using Mercuri sores. More information here
• Mercury: python package for processing and analysing Mercuri scores.

How can I get a MYO-Guide accout?

MYO-Guide is currently in a closed beta. This means that accounts are created manually under request. If you would like to try MYO-Guide, please contact [email protected]

Who is behid MYO-Guide

MYO-Guide has been developed in the John Walton Muscular Dystrophy Research Centre (JWMDRC), a world-leading referent in the research and care for neuromuscular diseases. The JWMDRC is part of Newcastle University and works hand-to-hand with the NHS.

Prof. Jordi Díaz-Manera

Jordi Díaz-Manera is Professor of Neuromuscular Diseases, Genetics and Translational research at John Walton Muscular Dystrophy Research Centre at Newcastle University in the UK. He is an adult neurologist specialized in neuromuscular diseases with more than 20 years experience in the diagnosis and follow-up of patients in clinics. Jordi has been working in implementing MRI as an useful tool for the diagnosis of inherited and acquired neuromuscular diseases for more than 10 years. He has written several papers describing the mean radiological features of diseases such as limb girdle muscular dystrophy R2, Pompe disease, FSHD type II, limb girdle muscular dystrophy produced by mutations in the LMNA gene. Moreover, Jordi is interested in quantitative MRI sequences that can be used to follow-up muscle degeneration with different neuromuscular diseases in natural history studies such as Dixon or T2.

José Verdú-Díaz, Eng.

José Verdú-Díaz is a Research Assistant at the John Walton Muscular Dystrophy Research Centre. He developed the first version of Myo-Guide during his Bachelor's in Networks Engineering under the coordination of Prof. Jordi Díaz-Manera. Since then, he has been collaborating on multiple projects involving Artificial Intelligence and biomedical data. In 2022, José started working in the JWMDRC to develop the second version of Myo-Guide and study how AI can be used to process and analyse MRI to obtain insights, diagnostics and prognostics of neuromuscular diseases. He helped in the creation of the dataset used to train the AI tools that MYO-Guide offers.

Carla Florencia Bolaño-Díaz, MD

Carla Bolaño-Diaz is a Clinical Research Associate at the John Walton Muscular } Dystrophy Research Centre, Newcastle University in the United Kingdom. She trained as an adult neurologist in Argentina and started showing a special interest in neuromuscular diseases while she was in training. She worked alongside Prof. Alberto Dubrovsky and once she finished her general neurology training, relocated to Newcastle to start working at the JWMDRC in order to specialize in genetic muscle conditions. She is studying and further developing the use of MRI as a biomarker and diagnostic, follow up and prognostic tool in dysferlinopathy patients as her PhD project.

Alejandro González Chamorro

Alejandro Gonzalez Chamorro is a Project Coordinator at the John Walton Muscular Dystrophy Research Centre at Newcastle University, UK. He works on a few local and international Basic and Translational Research projects, among which he coordinates the MYO-Guide. Alejandro has been working on the set-up of the study; coordinates the sites and collaborators from different countries; collates and manages the data; and acts as the first point of contact for the MYO-Share platform.

Jaume Bacardit

Prof. Jaume Bacardit is Professor of Artificial Intelligence (from August 2023) at the School of Computing of Newcastle University. Bacardit's research interests include the development of machine learning methods for large-scale problems, explainable AI, and the application of machine learning to life sciences real-world problems. He leads/has led the data analytics efforts of several large biological/biomedical interdisciplinary consortiums: APPROACH and PORTABOLOMICS. Interpreting how machine learning models take decisions has been one of Bacardit's active areas of research for many years, way before the term Explainable AI emerged.

Sam Fitzsimmons

Sam Fitzsimmons is a Project Manager at the John Walton Muscular Dystrophy Centre, Newcastle University, UK. Sam has over 5 years of experience as a Clinical Trial Coordinator, and started her research career in the North East and North Cumbria Clinical Research Network (NENC CRN) as the Industry Coordinator, supporting the facilitation of research in the North East. Sam project manages several basic and translational research projects within the centre and is responsible for project oversight and support, maintaining and managing the planning and risk processes, as well as budget management.

Acknowledgements

Holly Borland

Holly Borland is a medical student at Newcastle University working at the JWMDRC to assist with ongoing research and centre projects. She has a background in physiology and prior experience as a clinical trial coordinator in a neurology department. She is passionate about the integration of research and clinical care for patients and hopes to continue contributing to these two realms as I work towards becoming a doctor.

Darryl Darian

Darryl Darian is an intercalating medical student from Universitas Indonesia currently taking the M.Res programme in Newcastle University. He discovered his interest in research while taking his bachelor's degree and yearns to contribute to the development of artificial intelligence in the world of medicine. He has been working on scoring fat infiltration in MRI to create the dataset for the diagnosis model neuromuscular disease.

Shona Haston

Shona Haston is a Research Assistant Intern at the John Walton Muscular Dystrophy Research Centre. She is a PhD student at Newcastle University, researching orthopaedic implants. She has been working on segmenting muscle regions of MRIs, which will be used to train the next version of MYO-Guide.

Elena Stebbings

Elena contributed to the annotation of the training dataset for MyoGuide. She has recently earned her Bachelor of Science degrees in Psychology and Biology and has begun collaborative work with the Neurobiology and Experimental Therapeutics Group, led by Assistant Professor Phil Burnet, as well as with the Oxford Centre for Microbiome Studies, alongside Jethro Johnson.

Debora Alvim

Debora Mucida Alvim, a fourth-year medical student at Newcastle University, is presently serving as a Research Assistant Intern at the John Walton Muscular Dystrophy Research Centre. She has played a crucial role in the MyoGuide team by annotating the training data for both the segmentation and diagnosis models. Debora is interested on exploring how machine learning can transform diagnostic methods and enhance patient care.

Citing MYO-Guide

If you are using MYO-Guide as part of your research, you can use one of the following:

Article

Accuracy of a machine learning muscle MRI-based tool for the diagnosis of muscular dystrophies

José Verdú-Díaz, Jorge Alonso-Pérez, Claudia Nuñez-Peralta, Giorgio Tasca, John Vissing, Volker Straub, Roberto Fernández-Torrón, Jaume Llauger, Isabel Illa, Jordi Díaz-Manera

First published February 6, 2020, DOI: https://doi.org/10.1212/WNL.0000000000009068