EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

CODA: an open-source platform for federated analysis and machine learning on distributed healthcare data

Louis Mullie iD, Jonathan Afilalo iD, Patrick Archambault, Rima Bouchakri, Kip Brown, David L Buckeridge iD, Yiorgos Alexandros Cavayas, Alexis F Turgeon, Denis Martineau, François Lamontagne, Martine Lebrasseur, Renald Lemieux und 6 weitere

Journal of the American Medical Informatics Association · 16.02.2024 · 31 (3) · 651-665

DOI: 10.1093/jamia/ocad235 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of the American Medical Informatics Association, online erschienen am 21.12.2023. Verfasst von Louis Mullie und 17 weiteren Autor:innen. Beteiligte Einrichtung: Department of Medicine, Centre Hospitalier de l'Université de Montréal, Montréal, H2X 3E4, Canada. Hauptthemen laut MEDLINE: Software, Health Facilities. Die Arbeit wurde bisher 24-mal zitiert (Crossref) und nennt 55 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: CIHR, Canadian Institutes of Health Research, Québec Table Nationale des Directeurs de Recherche.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Objectives

Distributed computations facilitate multi-institutional data analysis while avoiding the costs and complexity of data pooling. Existing approaches lack crucial features, such as built-in medical standards and terminologies, no-code data visualizations, explicit disclosure control mechanisms, and support for basic statistical computations, in addition to gradient-based optimization capabilities. Materials and methods We describe the development of the Collaborative Data Analysis (CODA) platform, and the design choices undertaken to address the key needs identified during our survey of stakeholders. We use a public dataset (MIMIC-IV) to demonstrate end-to-end multi-modal FL using CODA. We assessed the technical feasibility of deploying the CODA platform at 9 hospitals in Canada, describe implementation challenges, and evaluate its scalability on large patient populations.

Results

The CODA platform was designed, developed, and deployed between January 2020 and January 2023. Software code, documentation, and technical documents were released under an open-source license. Multi-modal federated averaging is illustrated using the MIMIC-IV and MIMIC-CXR datasets. To date, 8 out of the 9 participating sites have successfully deployed the platform, with a total enrolment of >1M patients. Mapping data from legacy systems to FHIR was the biggest barrier to implementation. Discussion and conclusion The CODA platform was developed and successfully deployed in a public healthcare setting in Canada, with heterogeneous information technology systems and capabilities. Ongoing efforts will use the platform to develop and prospectively validate models for risk assessment, proactive monitoring, and resource usage. Further work will also make tools available to facilitate migration from legacy formats to FHIR and DICOM.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

SoftwareHealth FacilitiesHumansDelivery of Health CareMachine LearningCanada

Stichwörter der Autor:innen

biomedical analyticsdistributed computingfederated learninghealthcare data managementmachine learningpredictive modelsresource usage analysis

Autor:innen und Institutionen

  • Department of Medicine, Centre Hospitalier de l'Université de Montréal, Montréal, H2X 3E4, Canada.
  • Faculty of Medicine, Université de Montréal, Montréal, H3C 3J7, Canada.
  • Mila Quebec Artificial Intelligence Institute, Montréal, H2S 3H1, Canada.
  • Department of Medicine, Jewish General Hospital, Montréal, H3T 1E4, Canada.
  • Department of Emergency Medicine and Family Medicine, Université Laval, Québec, G1V 0A6, Canada.
  • Department of Anesthesiology and Critical Care Medicine, Université Laval, Québec, G1V 0A6, Canada.

ORCID-Profile: Louis Mullie, Jonathan Afilalo, David L Buckeridge

Förderung

  • CIHR
  • Canadian Institutes of Health Research
  • Québec Table Nationale des Directeurs de Recherche
  • Réseau de bio-imagerie du Québec
  • Fondation du Centre Hospitalier de l'Université de Montréal
  • CHUM Department of Critical Care
  • Fonds de Recherche du Québec Santé
  • Canada Research Chair in Critical Care Neurology and Trauma

Zitieren

APA

Mullie, L., Afilalo, J., Archambault, P., Bouchakri, R., Brown, K., Buckeridge, D. L., Cavayas, Y. A., Turgeon, A. F., Martineau, D., Lamontagne, F., Lebrasseur, M., Lemieux, R., Li, J., Sauthier, M., St-Onge, P., Tang, A., Witteman, W., & Chassé, M. (2024). CODA: an open-source platform for federated analysis and machine learning on distributed healthcare data. Journal of the American Medical Informatics Association, 31(3), 651-665. https://doi.org/10.1093/jamia/ocad235

Vancouver

Mullie L, Afilalo J, Archambault P, Bouchakri R, Brown K, Buckeridge DL, et al. CODA: an open-source platform for federated analysis and machine learning on distributed healthcare data. Journal of the American Medical Informatics Association. 2024;31(3):651-665. doi:10.1093/jamia/ocad235

BibTeX
@article{Mullie2024,
  author = {Mullie, Louis and Afilalo, Jonathan and Archambault, Patrick and Bouchakri, Rima and Brown, Kip and Buckeridge, David L and Cavayas, Yiorgos Alexandros and Turgeon, Alexis F and Martineau, Denis and Lamontagne, François and Lebrasseur, Martine and Lemieux, Renald and Li, Jeffrey and Sauthier, Michaël and St-Onge, Pascal and Tang, An and Witteman, William and Chassé, Michaël},
  title = {CODA: an open-source platform for federated analysis and machine learning on distributed healthcare data},
  journal = {Journal of the American Medical Informatics Association},
  year = {2024},
  volume = {31},
  number = {3},
  pages = {651--665},
  doi = {10.1093/jamia/ocad235}
}
Kostenlos starten →