Preferred Label : Federated Learning;
CISMeF synonym : collaborative learning;
MeSH synonym : Federated Learning, Machine; Federated Learnings; Federated Machine Learning; Federated Machine Learnings; Learning, Federated; Machine Federated Learning; Machine Federated Learnings; Machine Learning, Federated;
Définition CISMeF : A decentralized approach to training ML models. Models are trained by each site on
data that are kept locally, and model updates are sent to a central server, whereby
the central server aggregates these updates to improve a global model. This method
is designed to preserve data privacy, as raw data remain at the local sites and are
not centralized. For example, federated learning can allow hospitals to collaborate
on a heart disease prediction model without sharing patient data. The model is sent
to be trained locally at each hospital, and only the model updates from each hospital,
not raw data, are sent back and aggregated. This way, individual patient information
remains localized, addressing privacy concerns while still benefiting from a collectively
improved model.; Source: Adapted from: Darzidehkalani, E., Ghasemi-rad, M., & van Ooijen, P. M. A.
(2022). Federated Learning in Medical Imaging: Part I: Toward Multicentral Health
Care Ecosystems. Journal of the American College of Radiology, 19(8), 969–974. https://doi.org/10.1016/j.jacr.2022.03.015External
Link Disclaimer IEEE Standards. (2022). IEEE Standard for Performance and Safety Evaluation
of Artificial Intelligence Based Medical Devices: Terminology (IEEE Std 2802 ‐2022).
https://standards.ieee.org/ieee/2802/7460/;
Origin ID : 300271;
Automatic exact mappings (from CISMeF team)
- Currated CISMeF NLP mapping
A decentralized approach to training ML models. Models are trained by each site on
data that are kept locally, and model updates are sent to a central server, whereby
the central server aggregates these updates to improve a global model. This method
is designed to preserve data privacy, as raw data remain at the local sites and are
not centralized. For example, federated learning can allow hospitals to collaborate
on a heart disease prediction model without sharing patient data. The model is sent
to be trained locally at each hospital, and only the model updates from each hospital,
not raw data, are sent back and aggregated. This way, individual patient information
remains localized, addressing privacy concerns while still benefiting from a collectively
improved model.
Source: Adapted from: Darzidehkalani, E., Ghasemi-rad, M., & van Ooijen, P. M. A.
(2022). Federated Learning in Medical Imaging: Part I: Toward Multicentral Health
Care Ecosystems. Journal of the American College of Radiology, 19(8), 969–974. https://doi.org/10.1016/j.jacr.2022.03.015External
Link Disclaimer IEEE Standards. (2022). IEEE Standard for Performance and Safety Evaluation
of Artificial Intelligence Based Medical Devices: Terminology (IEEE Std 2802 ‐2022).
https://standards.ieee.org/ieee/2802/7460/
