Privacy-Aware Intelligent Healthcare Services with Federated Learning Architecture and Reinforcement Learning Agent

Lecture notes in electrical engineering(2023)

Cited 1|Views0
No score
The convergence of digital health systems, Internet of Healthcare Things (IoHT), and Deep Learning (DL) offers intelligent healthcare services to deploy in medical software development with extended decision-making modules. However, the communication reliability and information privacy of data-driven model integration remain a challenging topic to further discuss by researchers and standard organizations. From a real-time communication perspective, the standard of Quality-of-Service requirements in smart healthcare is within real-time services, which permits extremely low packet delay budget and error loss rate. To jointly generate a delay-aware, reliability-aware, and privacy-aware approaches in intelligent healthcare services (e.g. DL-based medical software), there are three key essential aspects, namely distributed edge learning, resource virtualization, and model slicing orchestration. This paper presents optimized Federated Learning (FL) architecture based on autonomous agent policies (resource placement and slicing orchestration) using reinforcement learning. The system architecture outlines large-scale IoHT deployment in wireless cellular networks, network functions virtualization-enabled edge computing placement, and converged FL components. The communication and computation models tackle the utilization efficiencies of bandwidth, energy, and computing capacities. This proposed optimization approach strengthens privacy and reliability by orchestrating through different weights of slicing prioritization in multi-intelligent healthcare services such as non-real-time/real-time modelling, telemedicine, and remote operation (e.g. telesurgery).
Translated text
Key words
federated learning architecture,federated learning,healthcare,privacy-aware
AI Read Science
Must-Reading Tree
Generate MRT to find the research sequence of this paper
Chat Paper
Summary is being generated by the instructions you defined