Peer-to-Peer Federated Learning for COVID-19 Detection Using Transformers
The simultaneous advances in deep learning and the Internet of Things (IoT) have benefited distributed deep learning paradigms. Federated learning is one of the most promising frameworks, where a server works with local learners to train a global model. The intrinsic heterogeneity of IoT devices, or...
Main Authors: | , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2023-05-01
|
Series: | Computers |
Subjects: | |
Online Access: | https://www.mdpi.com/2073-431X/12/5/106 |
_version_ | 1797600538411401216 |
---|---|
author | Mohamed Chetoui Moulay A. Akhloufi |
author_facet | Mohamed Chetoui Moulay A. Akhloufi |
author_sort | Mohamed Chetoui |
collection | DOAJ |
description | The simultaneous advances in deep learning and the Internet of Things (IoT) have benefited distributed deep learning paradigms. Federated learning is one of the most promising frameworks, where a server works with local learners to train a global model. The intrinsic heterogeneity of IoT devices, or non-independent and identically distributed (Non-I.I.D.) data, combined with the unstable communication network environment, causes a bottleneck that slows convergence and degrades learning efficiency. Additionally, the majority of weight averaging-based model aggregation approaches raise questions about learning fairness. In this paper, we propose a peer-to-peer federated learning (P2PFL) framework based on Vision Transformers (ViT) models to help solve some of the above issues and classify COVID-19 vs. normal cases on Chest-X-Ray (CXR) images. Particularly, clients jointly iterate and aggregate the models in order to build a robust model. The experimental results demonstrate that the proposed approach is capable of significantly improving the performance of the model with an Area Under Curve (AUC) of 0.92 and 0.99 for hospital-1 and hospital-2, respectively. |
first_indexed | 2024-03-11T03:49:32Z |
format | Article |
id | doaj.art-3ecf34e96fe145d189ab60242f6b68f4 |
institution | Directory Open Access Journal |
issn | 2073-431X |
language | English |
last_indexed | 2024-03-11T03:49:32Z |
publishDate | 2023-05-01 |
publisher | MDPI AG |
record_format | Article |
series | Computers |
spelling | doaj.art-3ecf34e96fe145d189ab60242f6b68f42023-11-18T00:58:30ZengMDPI AGComputers2073-431X2023-05-0112510610.3390/computers12050106Peer-to-Peer Federated Learning for COVID-19 Detection Using TransformersMohamed Chetoui0Moulay A. Akhloufi1Perception, Robotics, and Intelligent Machines Research Group (PRIME), Department of Computer Science, Université de Moncton, Moncton, NB E1A 3E9, CanadaPerception, Robotics, and Intelligent Machines Research Group (PRIME), Department of Computer Science, Université de Moncton, Moncton, NB E1A 3E9, CanadaThe simultaneous advances in deep learning and the Internet of Things (IoT) have benefited distributed deep learning paradigms. Federated learning is one of the most promising frameworks, where a server works with local learners to train a global model. The intrinsic heterogeneity of IoT devices, or non-independent and identically distributed (Non-I.I.D.) data, combined with the unstable communication network environment, causes a bottleneck that slows convergence and degrades learning efficiency. Additionally, the majority of weight averaging-based model aggregation approaches raise questions about learning fairness. In this paper, we propose a peer-to-peer federated learning (P2PFL) framework based on Vision Transformers (ViT) models to help solve some of the above issues and classify COVID-19 vs. normal cases on Chest-X-Ray (CXR) images. Particularly, clients jointly iterate and aggregate the models in order to build a robust model. The experimental results demonstrate that the proposed approach is capable of significantly improving the performance of the model with an Area Under Curve (AUC) of 0.92 and 0.99 for hospital-1 and hospital-2, respectively.https://www.mdpi.com/2073-431X/12/5/106federated learningVision Transformersdeep learningCOVID-19medical imaging |
spellingShingle | Mohamed Chetoui Moulay A. Akhloufi Peer-to-Peer Federated Learning for COVID-19 Detection Using Transformers Computers federated learning Vision Transformers deep learning COVID-19 medical imaging |
title | Peer-to-Peer Federated Learning for COVID-19 Detection Using Transformers |
title_full | Peer-to-Peer Federated Learning for COVID-19 Detection Using Transformers |
title_fullStr | Peer-to-Peer Federated Learning for COVID-19 Detection Using Transformers |
title_full_unstemmed | Peer-to-Peer Federated Learning for COVID-19 Detection Using Transformers |
title_short | Peer-to-Peer Federated Learning for COVID-19 Detection Using Transformers |
title_sort | peer to peer federated learning for covid 19 detection using transformers |
topic | federated learning Vision Transformers deep learning COVID-19 medical imaging |
url | https://www.mdpi.com/2073-431X/12/5/106 |
work_keys_str_mv | AT mohamedchetoui peertopeerfederatedlearningforcovid19detectionusingtransformers AT moulayaakhloufi peertopeerfederatedlearningforcovid19detectionusingtransformers |