Paper Title: “TrustFedHealth: Federated Learning with Homomorphic Encryption and Blockchain for Heart Disease Prediction in Smart Healthcare”
Student Contributors:
- Bùi Tấn Hải Đăng (Talent - Information Security 2020)
- Phan Hữu Luân (High Quality - Information Security 2020.1)
- Vương Đinh Thanh Ngân (High Quality - Information Security ATCL2020.1)
Supervisors:
- TS. Phạm Văn Hậu
- ThS Phan Thế Duy
Paper Summary: In recent years, advances in the Internet of Medical Things (IoMT) or smart devices have enabled automatic human health monitoring. The use of smart healthcare devices can not only reduce the burden on hospitals but also save costs, travel time, and provide home-based disease diagnosis, such as stroke rates. IoMT generates a large amount of data that can be used to train machine learning (ML) models for accurate disease diagnosis. However, data used to train ML models is often privacy-preserving, making sharing difficult and requiring high security. To overcome this challenge, this paper proposes a collaborative framework called TrustFedHealth to train heart disease prediction models based on ML. TrustFedHealth uses Federated Learning (FL) to enable training with decentralized data stored separately on multiple machines. Additionally, Mobile Edge Computing (MEC) is integrated into the model as a solution to optimize communication time between different elements of the system and reduce congestion. Homomorphic Encryption (HE) is also combined with FL to protect the security of model updates between client and aggregate server. Furthermore, Blockchain (BC) is leveraged to address the traceability of contributions and ensure the transparency of model updates. Through evaluation results on the MIT-BIH Arrhythmia Database from Physionet, TrustFedHealth provides a promising method for training ML-based heart disease prediction models while maintaining privacy and security.
"We would like to express our gratitude to Dr. Phạm Văn Hậu and Mr. Phan Thế Duy for their dedicated support and assistance throughout our research."
The International Conference on Advanced Technologies for Communications (ATC) is an annual scientific conference specializing in electronics, communications, information technology, and related fields. The conference opens up opportunities for exchange, research presentation, technology transfer, and introduction of products from businesses operating in these fields. The conference has been held annually since 2008, attracting over 200 papers with about 150 scientists from many countries worldwide. ATC 2023 was hosted by the University of Information Technology and Communication Vietnam - Korea and took place in Da Nang City, Vietnam from 19-21/10/2023.
For more details, visit: https://www.facebook.com/UIT.Fanpage/posts/pfbid02R4eK85sSVFrnAoRPfZrtzsgihuaWYXvK7QJW5dvRMWi8WptdUfT1ow7rwebL3ArAl
Written by: Hai Bang
Translated by: Ngoc Diem