Federated Learning For Heterogeneous Electronic Health Records Utilising Augmented Temporal Graph Attention Networks

Soheila Molaei, Anshul Thakur, Ghazaleh Niknam, Andrew Soltan, Hadi Zare, David A Clifton
Proceedings of The 27th International Conference on Artificial Intelligence and Statistics, PMLR 238:1342-1350, 2024.

Abstract

The proliferation of decentralised electronic healthcare records (EHRs) across medical institutions requires innovative federated learning strategies for collaborative data analysis and global model training, prioritising data privacy. A prevalent issue during decentralised model training is the data-view discrepancies across medical institutions that arises from differences or availability of healthcare services, such as blood test panels. The prevailing way to handle this issue is to select a common subset of features across institutions to make data-views consistent. This approach, however, constrains some institutions to shed some critical features that may play a significant role in improving the model performance. This paper introduces a federated learning framework that relies on augmented graph attention networks to address data-view heterogeneity. The proposed framework utilises an alignment augmentation layer over self-attention mechanisms to weigh the importance of neighbouring nodes when updating a node’s embedding irrespective of the data-views. Furthermore, our framework adeptly addresses both the temporal nuances and structural intricacies of EHR datasets. This dual capability not only offers deeper insights but also effectively encapsulates EHR graphs’ time-evolving nature. Using diverse real-world datasets, we show that the proposed framework significantly outperforms conventional FL methodology for dealing with heterogeneous data-views.

Cite this Paper

BibTeX
@InProceedings{pmlr-v238-molaei24a, title = {Federated Learning For Heterogeneous Electronic Health Records Utilising Augmented Temporal Graph Attention Networks}, author = {Molaei, Soheila and Thakur, Anshul and Niknam, Ghazaleh and Soltan, Andrew and Zare, Hadi and A Clifton, David}, booktitle = {Proceedings of The 27th International Conference on Artificial Intelligence and Statistics}, pages = {1342--1350}, year = {2024}, editor = {Dasgupta, Sanjoy and Mandt, Stephan and Li, Yingzhen}, volume = {238}, series = {Proceedings of Machine Learning Research}, month = {02--04 May}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v238/molaei24a/molaei24a.pdf}, url = {https://proceedings.mlr.press/v238/molaei24a.html}, abstract = {The proliferation of decentralised electronic healthcare records (EHRs) across medical institutions requires innovative federated learning strategies for collaborative data analysis and global model training, prioritising data privacy. A prevalent issue during decentralised model training is the data-view discrepancies across medical institutions that arises from differences or availability of healthcare services, such as blood test panels. The prevailing way to handle this issue is to select a common subset of features across institutions to make data-views consistent. This approach, however, constrains some institutions to shed some critical features that may play a significant role in improving the model performance. This paper introduces a federated learning framework that relies on augmented graph attention networks to address data-view heterogeneity. The proposed framework utilises an alignment augmentation layer over self-attention mechanisms to weigh the importance of neighbouring nodes when updating a node’s embedding irrespective of the data-views. Furthermore, our framework adeptly addresses both the temporal nuances and structural intricacies of EHR datasets. This dual capability not only offers deeper insights but also effectively encapsulates EHR graphs’ time-evolving nature. Using diverse real-world datasets, we show that the proposed framework significantly outperforms conventional FL methodology for dealing with heterogeneous data-views.} }
Endnote
%0 Conference Paper %T Federated Learning For Heterogeneous Electronic Health Records Utilising Augmented Temporal Graph Attention Networks %A Soheila Molaei %A Anshul Thakur %A Ghazaleh Niknam %A Andrew Soltan %A Hadi Zare %A David A Clifton %B Proceedings of The 27th International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2024 %E Sanjoy Dasgupta %E Stephan Mandt %E Yingzhen Li %F pmlr-v238-molaei24a %I PMLR %P 1342--1350 %U https://proceedings.mlr.press/v238/molaei24a.html %V 238 %X The proliferation of decentralised electronic healthcare records (EHRs) across medical institutions requires innovative federated learning strategies for collaborative data analysis and global model training, prioritising data privacy. A prevalent issue during decentralised model training is the data-view discrepancies across medical institutions that arises from differences or availability of healthcare services, such as blood test panels. The prevailing way to handle this issue is to select a common subset of features across institutions to make data-views consistent. This approach, however, constrains some institutions to shed some critical features that may play a significant role in improving the model performance. This paper introduces a federated learning framework that relies on augmented graph attention networks to address data-view heterogeneity. The proposed framework utilises an alignment augmentation layer over self-attention mechanisms to weigh the importance of neighbouring nodes when updating a node’s embedding irrespective of the data-views. Furthermore, our framework adeptly addresses both the temporal nuances and structural intricacies of EHR datasets. This dual capability not only offers deeper insights but also effectively encapsulates EHR graphs’ time-evolving nature. Using diverse real-world datasets, we show that the proposed framework significantly outperforms conventional FL methodology for dealing with heterogeneous data-views.
APA
Molaei, S., Thakur, A., Niknam, G., Soltan, A., Zare, H. & A Clifton, D.. (2024). Federated Learning For Heterogeneous Electronic Health Records Utilising Augmented Temporal Graph Attention Networks. Proceedings of The 27th International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 238:1342-1350 Available from https://proceedings.mlr.press/v238/molaei24a.html.

Related Material