Making head and neck cancer clinical data Findable-Accessible-Interoperable-Reusable to support multi-institutional collaboration and federated learning
BJR|Artificial Intelligence · 04.03.2024 · 1 (1)
DOI: 10.1093/bjrai/ubae005 ↗Kurz gefasst
Fachartikel in der Zeitschrift BJR|Artificial Intelligence, online erschienen am 06.03.2024. Verfasst von Varsha Gouthamchand und 10 weiteren Autor:innen. Beteiligte Einrichtung: Clinical Data Science, Faculty of Health Medicine and Life Sciences, Maastricht University, Maastricht 6229 EN, The Netherlands. Die Arbeit wurde bisher 7-mal zitiert (Crossref) und nennt 43 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Dutch Research Council, Hanarth Foundation.
Abstract (Originalsprache)
Objectives
Federated learning (FL) is a group of methodologies where statistical modelling can be performed without exchanging identifiable patient data between cooperating institutions. To realize its potential for AI development on clinical data, a number of bottlenecks need to be addressed. One of these is making data Findable-Accessible-Interoperable-Reusable (FAIR). The primary aim of this work is to show that tools making data FAIR allow consortia to collaborate on privacy-aware data exploration, data visualization, and training of models on each other’s original data.
Methods
We propose a “Schema-on-Read” FAIR-ification method that adapts for different (re)analyses without needing to change the underlying original data. The procedure involves (1) decoupling the contents of the data from its schema and database structure, (2) annotation with semantic ontologies as a metadata layer, and (3) readout using semantic queries. Open-source tools are given as Docker containers to help local investigators prepare their data on-premises.
Results
We created a federated privacy-preserving visualization dashboard for case mix exploration of 5 distributed datasets with no common schema at the point of origin. We demonstrated robust and flexible prognostication model development and validation, linking together different data sources—clinical risk factors and radiomics.
Conclusions
Our procedure leads to successful (re)use of data in FL-based consortia without the need to impose a common schema at every point of origin of data. Advances in knowledge This work supports the adoption of FL within the healthcare AI community by sharing means to make data more FAIR.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Clinical Data Science, Faculty of Health Medicine and Life Sciences, Maastricht University, Maastricht 6229 EN, The Netherlands.
- Dept of Radiation Oncology (MAASTRO), School of Oncology and Reproduction, Maastricht University Medical Centre+, Maastricht 6229 ET, The Netherlands.
- Radiation Medicine Program, Princess Margaret Cancer Centre, University Health Network, Toronto, ON M5G 2C4, Canada.
- Cancer Digital Intelligence, Princess Margaret Cancer Centre, Toronto, ON M5G 2C4, Canada.
- Dept of Radiation Oncology, Greater Poland Cancer Centre II, Poznan 61-866, Poland.
- Medical Biophysics University of Toronto, Vector Institute for Artificial Intelligence and Ontario Institute for Cancer Research, Toronto, ON M5G 0C6, Canada.
ORCID-Profile: Leonard Wee
Förderung
- Dutch Research Council
- Hanarth Foundation
Zitieren
APA
Gouthamchand, V., Choudhury, A., Hoebers, F. J. P., Wesseling, F. W. R., Welch, M., Kim, S., Kazmierska, J., Dekker, A., Haibe-Kains, B., van Soest, J., & Wee, L. (2024). Making head and neck cancer clinical data Findable-Accessible-Interoperable-Reusable to support multi-institutional collaboration and federated learning. BJR|Artificial Intelligence, 1(1). https://doi.org/10.1093/bjrai/ubae005
Vancouver
Gouthamchand V, Choudhury A, Hoebers FJP, Wesseling FWR, Welch M, Kim S, et al. Making head and neck cancer clinical data Findable-Accessible-Interoperable-Reusable to support multi-institutional collaboration and federated learning. BJR|Artificial Intelligence. 2024;1(1). doi:10.1093/bjrai/ubae005
BibTeX
@article{Gouthamchand2024,
author = {Gouthamchand, Varsha and Choudhury, Ananya and Hoebers, Frank J P and Wesseling, Frederik W R and Welch, Mattea and Kim, Sejin and Kazmierska, Joanna and Dekker, Andre and Haibe-Kains, Benjamin and van Soest, Johan and Wee, Leonard},
title = {Making head and neck cancer clinical data Findable-Accessible-Interoperable-Reusable to support multi-institutional collaboration and federated learning},
journal = {BJR|Artificial Intelligence},
year = {2024},
volume = {1},
number = {1},
doi = {10.1093/bjrai/ubae005}
}
EUVIMED