Magnetic Resonance Imaging Sequence Identification Using a Metadata Learning Approach

Frontiers in Neuroinformatics

Christopher J. M. Scott¹  Raymond W. Lam²  The ONDRI Investigators⁵  Angela Troyer⁶  Benicio N. Frey⁷  Roumen Milev⁸  Daniel J. Müller^1,10  Sidney H. Kennedy^1,13  Sandra E. Black^1,15  Richard H. Swartz^1,15  Douglas P. Munoz^1,15  Manuel Montero-Odasso^1,15  Robert Bartha^1,15  Robert A. Hegele^1,15  Barry Greenberg^1,15  Chris Hudson^1,15  Maria Carmela Tartaglia^1,15  J. B. Orange^1,15  Paula M. McLaughlin^1,15  Lorne Zinman^1,15  Stephen C. Strother^1,15  Dale Corbett^1,15  Anthony E. Lang^1,15  Mario Masellis^1,15  Morris Freedman^1,15  David A. Grimes^1,15  Sean Symons^1,15  Elizabeth Finger^1,15  Michael Borrie^1,15  Michael J. Strong^1,15  William E. McIlroy^1,15  David G. Munoz^1,15  Evdokia Anagnostou^1,18  Glenda M. MacQueen^1,19  Stefanie Hassel^1,19  Shuai Liang^2,20  Tom Gee^2,20  Derek Beaton^2,23  Stephen R. Arnott^2,23  Mojdeh Zamyadi^2,23  Jason P. Lerch^2,24 

[1] Stroke Foundation Centre for Stroke Recovery, Toronto, ON, Canada;0Department of Psychiatry, University of British Columbia, Vancouver, BC, Canada;;0Heart &1Department of Psychiatry and Behavioral Neurosciences, McMaster University, Hamilton, ON, Canada;1Sunnybrook Health Sciences Centre, Brain Sciences Research Program, Sunnybrook Research Institute, Toronto, ON, Canada;2Department of Medical Biophysics, University of Toronto, Toronto, ON, Canada;2Mood Disorders Program, St. Joseph’s Healthcare, Hamilton, ON, Canada;3Departments of Psychiatry and Psychology, Providence Care Hospital, Queen’s University, Kingston, ON, Canada;4Molecular Brain Science, Centre for Addiction and Mental Health, Campbell Family Mental Health Research Institute, Toronto, ON, Canada;5Department of Psychiatry, University of Toronto, Toronto, ON, Canada;6Department of Psychiatry, Krembil Research Centre, University Health Network, Toronto, ON, Canada;7Department of Psychiatry, St. Michael’s Hospital, University of Toronto, Toronto, ON, Canada;8Keenan Research Centre for Biomedical Science, St. Michael’s Hospital, Li Ka Shing Knowledge Institute, Toronto, ON, Canada;9L.C. Campbell Cognitive Neurology Research Unit, Toronto, ON, Canada;;Bloorview Research Institute, Holland Bloorview Kids Rehabilitation Hospital, Toronto, ON, Canada;Department of Medical Imaging, Sunnybrook Health Sciences Centre, Toronto, ON, Canada;Department of Pediatrics, University of Toronto, Toronto, ON, Canada;Department of Psychiatry, Cumming School of Medicine, University of Calgary, Calgary, AB, Canada;Indoc Research, Toronto, ON, Canada;Mouse Imaging Centre, Hospital for Sick Children, Toronto, ON, Canada;Robarts Research Institute, Western University, London, ON, Canada;Rotman Research Institute, Baycrest Health Center, Toronto, ON, Canada;Wellcome Centre for Integrative Neuroimaging, FMRIB, Nuffield Department of Clinical Neurosciences, University of Oxford, Oxford, United Kingdom;

关键词: health data;    MRI sequence naming standardization;    data share and exchange;    machine learning;    metadata learning;    AI-assisted data management;   

DOI  :  10.3389/fninf.2021.622951

来源: DOAJ

【 摘 要 】

Despite the wide application of the magnetic resonance imaging (MRI) technique, there are no widely used standards on naming and describing MRI sequences. The absence of consistent naming conventions presents a major challenge in automating image processing since most MRI software require a priori knowledge of the type of the MRI sequences to be processed. This issue becomes increasingly critical with the current efforts toward open-sharing of MRI data in the neuroscience community. This manuscript reports an MRI sequence detection method using imaging metadata and a supervised machine learning technique. Three datasets from the Brain Center for Ontario Data Exploration (Brain-CODE) data platform, each involving MRI data from multiple research institutes, are used to build and test our model. The preliminary results show that a random forest model can be trained to accurately identify MRI sequence types, and to recognize MRI scans that do not belong to any of the known sequence types. Therefore the proposed approach can be used to automate processing of MRI data that involves a large number of variations in sequence names, and to help standardize sequence naming in ongoing data collections. This study highlights the potential of the machine learning approaches in helping manage health data.

【 授权许可】

Unknown