A COVID-19 Auxiliary Diagnosis Based on Federated Learning and Blockchainopen access
- Authors
- Wang, Z.; Cai, L.; Zhang, X.; Choi, Chang; Su, X.
- Issue Date
- Aug-2022
- Publisher
- Hindawi Limited
- Citation
- Computational and Mathematical Methods in Medicine, v.2022
- Journal Title
- Computational and Mathematical Methods in Medicine
- Volume
- 2022
- URI
- https://scholarworks.bwise.kr/gachon/handle/2020.sw.gachon/89880
- DOI
- 10.1155/2022/7078764
- ISSN
- 1748-670X
1748-6718
- Abstract
- Due to the high transmission rate and high pathogenicity of the novel coronavirus (COVID-19), there is an urgent need for the diagnosis and treatment of outbreaks around the world. In order to diagnose quickly and accurately, an auxiliary diagnosis method is proposed for COVID-19 based on federated learning and blockchain, which can quickly and effectively enable collaborative model training among multiple medical institutions. It is beneficial to address data sharing difficulties and issues of privacy and security. This research mainly includes the following sectors: in order to address insufficient medical data and the data silos, this paper applies federated learning to COVID-19's medical diagnosis to achieve the transformation and refinement of big data values. With regard to third-party dependence, blockchain technology is introduced to protect sensitive information and safeguard the data rights of medical institutions. To ensure the model's validity and applicability, this paper simulates realistic situations based on a real COVID-19 dataset and analyses problems such as model iteration delays. Experimental results demonstrate that this method achieves a multiparty participation in training and a better data protection and would help medical personnel diagnose coronavirus disease more effectively. © 2022 Ziyu Wang et al.
- Files in This Item
- There are no files associated with this item.
- Appears in
Collections - ETC > 1. Journal Articles
![qrcode](https://api.qrserver.com/v1/create-qr-code/?size=55x55&data=https://scholarworks.bwise.kr/gachon/handle/2020.sw.gachon/89880)
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.