HarmonyTM: multi-center data harmonization applied to distributed learning for Parkinson’s disease classification
HarmonyTM is a method tailored for harmonizing 3D neuroimaging data within the TM approach, aiming to minimize shortcut learning in distributed setups. This prevents the disease classifier from leveraging scanner-specific details to classify patients with or without PD-a key aspect for deploying ML models for clinical applications.
Authors
- Zahinoor Ismail (ORCID: https://orcid.org/0000-0002-5529-3731)
- Jasmine A. Moore (ORCID: https://orcid.org/0000-0001-9333-2730)
- Raissa Souza (ORCID: https://orcid.org/0000-0002-7455-3383)
- Oury Monchi (ORCID: https://orcid.org/0000-0003-2148-5409)
- Emma A. M. Stanley (ORCID: https://orcid.org/0000-0002-7802-6820)
- Nils D. Forkert (ORCID: https://orcid.org/0000-0003-2556-3224)
- Chris Kang
- Matthias Wilms (ORCID: https://orcid.org/0000-0001-8845-360X)
- Richard Camicioli (ORCID: https://orcid.org/0000-0003-2977-8660)
- Vedant Gulve
Institutions
- Indian Institute of Technology Kharagpur (IN)
- University of Alberta (CA)
- University of Calgary (CA)
- Women and Children’s Health Research Institute (CA)
- Hotchkiss Brain Institute (CA)
Publication Details
- Journal
- Journal of Medical Imaging
- Published
- 2024-09-20
- DOI
- https://doi.org/10.1117/1.jmi.11.5.054502
- Citations
- 1
- Primary Topic
- Parkinson's Disease Mechanisms and Treatments
- Type
- article
- Field-Weighted Citation Impact
- 0.37
Funders
- Alzheimer's Disease Neuroimaging Initiative
- Consortium canadien en neurodégénérescence associée au vieillissement
- Canada Research Chairs
- Parkinson Vereniging
- Institut de Valorisation des Données
- Parkinson Association of Alberta
- Natural Sciences and Engineering Research Council of Canada