Federated Learning in Distributed Medical Databases: Meta-Analysis of Large-Scale Subcortical Brain Data

By R. SantiagoS.Silva, Boris Gutman, E. Romero, Paul M. Thompson, André Altmann, Marco Lorenzi. 2018

204 citations recorded by the source corpus.

At this moment, databanks worldwide contain brain images of previously unimaginable numbers. Combined with developments in data science, these massive data provide the potential to better understand the genetic underpinnings of brain diseases. However, different datasets, which are stored at different institutions, cannot always be shared directly due to privacy and legal concerns, thus limiting the full exploitation of big data in the study of brain disorders. Here we propose a federated learning framework for securely accessing and meta-analyzing any biomedical data without sharing individual information. We illustrate our framework by investigating brain structural relationships across diseases and clinical cohorts. The framework is first tested on synthetic data and then applied to multi-centric, multi-database studies including ADNI, PPMI, MIRIAD and UK Biobank, showing the potential of the approach for further applications in distributed analysis of multi-centric cohorts.

Subfields: Artificial Intelligence, Cognitive Neuroscience, Radiology, Nuclear Medicine and Imaging

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Cognitive Neuroscience

In Radiology, Nuclear Medicine and Imaging

Citations in the corpus

Federated Learning: Challenges, Methods, and Future Directions (2019, methodology): “Federated learning methods have been deployed by major service providers [11, 124], and play a critical role in supporting privacy-sensitive applications where the training data are distributed at the edge [e.g., 5, 46, 51, 89, 105, 127, 139].”
Federated Learning for Healthcare Informatics (2019, methodology): “Representation learning [93] PCA 10–100 ADNI, UK Biobank, PPMI, MIRIAD”
A Survey on Federated Learning: The Journey From Centralized to Distributed On-Site Learning and Beyond (2021, unknown): “A framework for FL is proposed in [68] for the analysis of biomedical data.”
FedHome: Cloud-Edge Based Personalized Federated Learning for In-Home Health Monitoring (2020, background): “For example, Silva et al. present a federated learning framework by investigating brain structural relationships across diseases and clinical cohorts without sharing individual information [12].”
FedDG: Federated Domain Generalization on Medical Image Segmentation via Episodic Learning in Continuous Frequency Space (2021, background): “With the advantage of privacy protection, FL has recently drawn increasing interests in medical image applications [4, 18, 22, 27, 45, 49, 51].”
Federated Learning for Healthcare: Systematic Review and Architecture Proposal (2022, methodology): “[27] apply the ENIGMA Shape Analysis for prediction of neurological diseases in general, Lu et al.”
A Survey on federated learning* (2020, background): “It has been put forward by Silva et al.[57] to extract features from magnetic resonance images (MRI) come from different medical centers.”
Shifting machine learning for healthcare from development to deployment and from models to data (2022, background): “…focused on empirically evaluating model-design parameters, and on logistical decisions to optimize model performance and overcome the unique implementation challenges of federated learning, such as bottlenecks in protecting privacy and in tackling the statistical heterogeneity of the data 75,76 .”
FLOP: Federated Learning on Medical Datasets using Partial Networks (2021, methodology): “The Federated Learning framework has been applied to many healthcare tasks, such as predicting heart-related hospitalizations [5] and understanding the genetic underpinnings of brain diseases [27].”
A review of applications in federated learning (2020, methodology): “Federated principal components analysis (fPCA) has been put forward by Silva et al., (2019) to extract features from magnetic resonance images (MRI) come from different medical centers.”
Throughput-Optimal Topology Design for Cross-Silo Federated Learning (2020, background): “While the first FL papers [72, 51] emphasized the cross-device setting, the cross-silo scenario has become popular for distributed training among banks [107], hospitals [19, 93, 69], pharmaceutical labs [67], and manufacturers [74].”