The last decade has witnessed tremendous progress in digital healthcare delivery. This has led to the collection or generation of huge amounts of data, some of which have been released to the public for researchers to advance and find new cure and treatments and develop smarter ways for healthcare delivery using machine learning (ML) and artificial intelligence (AI). It is expected that AI and ML will bring a paradigm shift in the future of next-generation healthcare systems. These next-generation healthcare systems will utilize AI and take advantage of inexpensive high-performance and cloud computing environments. However, machine learning and AI are data-hungry and data-driven technologies. This characteristic could potentially limit the utilization of AI in healthcare systems. The data challenge in digital healthcare delivery is about more than data availability and data quality. Healthcare data and medical data have specific security and privacy requirements that restrict ML adoption. Federated Learning (FL) is an ML paradigm that could address data security and privacy restrictions in healthcare. However, applying FL in practice in healthcare requires a rigorous engineering approach that balances ML prediction quality and patient security and privacy. In this project, we propose a secure framework to enable federated learning at scale in healthcare systems. Studies show that the successful implementation of FL in healthcare has significant potential for enabling precision medicine at a large-scale. FL in healthcare will help in accelerating response to pandemic and outbreaks.
Journal & magazine links
- American Scientist Online
- Computer Communication Review
- Elsevier Computer Science
- Annual Computer Security Applications Conference
- IEEE Computer Society
- IEEE Security & Privacy Magazine
- IEICE Trans Search System
- International Journal of Digital Evidence
- ScienceDirect – Computers & Security
- ACM TISSEC (Transactions on Information & System Security)