Federated GNNs for EEG-Based Stroke Assessment

Andrea Protani, Lorenzo Giusti, Albert Sund Aillet, Simona Sacco, Paolo Manganotti, Lucio Marinelli, Diogo Reis Santos, Pierpaolo Brutti, Pietro Caliandro, Luigi Serio
Proceedings of UniReps: the Second Edition of the Workshop on Unifying Representations in Neural Models, PMLR 285:55-68, 2024.

Abstract

Machine learning (ML) has the potential to become an essential tool in supporting clinical decision-making processes, offering enhanced diagnostic capabilities and personalized treatment plans. However, outsourcing medical records to train ML models using patient data raises legal, privacy, and security concerns. Federated learning has emerged as a promising paradigm for collaborative ML, meeting healthcare institutions’ requirements for robust models without sharing sensitive data and compromising patient privacy. This study proposes a novel method that combines federated learning (FL) and Graph Neural Networks (GNNs) to predict stroke severity using electroencephalography (EEG) signals across multiple medical institutions. Our approach enables multiple hospitals to jointly train a shared GNN model on their local EEG data without exchanging patient information. Specifically, we address a regression problem by predicting the National Institutes of Health Stroke Scale (NIHSS), a key indicator of stroke severity. The proposed model leverages a masked self-attention mechanism to capture salient brain connectivity patterns and employs EdgeSHAP to provide post-hoc explanations of the neurological states after a stroke. We evaluated our method on EEG recordings from four institutions, achieving a mean absolute error (MAE) of 3.23 in predicting NIHSS, close to the average error made by human experts (MAE $\approx$ 3.0). This demonstrates the method’s effectiveness in providing accurate and explainable predictions while maintaining data privacy.

Cite this Paper

BibTeX
@InProceedings{pmlr-v285-protani24a, title = {Federated {GNN}s for {EEG}-Based Stroke Assessment}, author = {Protani, Andrea and Giusti, Lorenzo and Aillet, Albert Sund and Sacco, Simona and Manganotti, Paolo and Marinelli, Lucio and Santos, Diogo Reis and Brutti, Pierpaolo and Caliandro, Pietro and Serio, Luigi}, booktitle = {Proceedings of UniReps: the Second Edition of the Workshop on Unifying Representations in Neural Models}, pages = {55--68}, year = {2024}, editor = {Fumero, Marco and Domine, Clementine and Lähner, Zorah and Crisostomi, Donato and Moschella, Luca and Stachenfeld, Kimberly}, volume = {285}, series = {Proceedings of Machine Learning Research}, month = {14 Dec}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v285/main/assets/protani24a/protani24a.pdf}, url = {https://proceedings.mlr.press/v285/protani24a.html}, abstract = {Machine learning (ML) has the potential to become an essential tool in supporting clinical decision-making processes, offering enhanced diagnostic capabilities and personalized treatment plans. However, outsourcing medical records to train ML models using patient data raises legal, privacy, and security concerns. Federated learning has emerged as a promising paradigm for collaborative ML, meeting healthcare institutions’ requirements for robust models without sharing sensitive data and compromising patient privacy. This study proposes a novel method that combines federated learning (FL) and Graph Neural Networks (GNNs) to predict stroke severity using electroencephalography (EEG) signals across multiple medical institutions. Our approach enables multiple hospitals to jointly train a shared GNN model on their local EEG data without exchanging patient information. Specifically, we address a regression problem by predicting the National Institutes of Health Stroke Scale (NIHSS), a key indicator of stroke severity. The proposed model leverages a masked self-attention mechanism to capture salient brain connectivity patterns and employs EdgeSHAP to provide post-hoc explanations of the neurological states after a stroke. We evaluated our method on EEG recordings from four institutions, achieving a mean absolute error (MAE) of 3.23 in predicting NIHSS, close to the average error made by human experts (MAE $\approx$ 3.0). This demonstrates the method’s effectiveness in providing accurate and explainable predictions while maintaining data privacy.} }
Endnote
%0 Conference Paper %T Federated GNNs for EEG-Based Stroke Assessment %A Andrea Protani %A Lorenzo Giusti %A Albert Sund Aillet %A Simona Sacco %A Paolo Manganotti %A Lucio Marinelli %A Diogo Reis Santos %A Pierpaolo Brutti %A Pietro Caliandro %A Luigi Serio %B Proceedings of UniReps: the Second Edition of the Workshop on Unifying Representations in Neural Models %C Proceedings of Machine Learning Research %D 2024 %E Marco Fumero %E Clementine Domine %E Zorah Lähner %E Donato Crisostomi %E Luca Moschella %E Kimberly Stachenfeld %F pmlr-v285-protani24a %I PMLR %P 55--68 %U https://proceedings.mlr.press/v285/protani24a.html %V 285 %X Machine learning (ML) has the potential to become an essential tool in supporting clinical decision-making processes, offering enhanced diagnostic capabilities and personalized treatment plans. However, outsourcing medical records to train ML models using patient data raises legal, privacy, and security concerns. Federated learning has emerged as a promising paradigm for collaborative ML, meeting healthcare institutions’ requirements for robust models without sharing sensitive data and compromising patient privacy. This study proposes a novel method that combines federated learning (FL) and Graph Neural Networks (GNNs) to predict stroke severity using electroencephalography (EEG) signals across multiple medical institutions. Our approach enables multiple hospitals to jointly train a shared GNN model on their local EEG data without exchanging patient information. Specifically, we address a regression problem by predicting the National Institutes of Health Stroke Scale (NIHSS), a key indicator of stroke severity. The proposed model leverages a masked self-attention mechanism to capture salient brain connectivity patterns and employs EdgeSHAP to provide post-hoc explanations of the neurological states after a stroke. We evaluated our method on EEG recordings from four institutions, achieving a mean absolute error (MAE) of 3.23 in predicting NIHSS, close to the average error made by human experts (MAE $\approx$ 3.0). This demonstrates the method’s effectiveness in providing accurate and explainable predictions while maintaining data privacy.
APA
Protani, A., Giusti, L., Aillet, A.S., Sacco, S., Manganotti, P., Marinelli, L., Santos, D.R., Brutti, P., Caliandro, P. & Serio, L.. (2024). Federated GNNs for EEG-Based Stroke Assessment. Proceedings of UniReps: the Second Edition of the Workshop on Unifying Representations in Neural Models, in Proceedings of Machine Learning Research 285:55-68 Available from https://proceedings.mlr.press/v285/protani24a.html.

Related Material