AI-Enabled Healthcare and Enhanced Computational Resource Management With Digital Twins Into Task Offloading Strategies

Abstract

Efficient management of computational resources and data in the healthcare sector is increasingly challenging, particularly with the advent of advanced healthcare technologies. Effective task offloading mechanisms are crucial for enhancing system performance, patient care, and data security. This study aims to introduce and evaluate a novel framework for task offloading in healthcare environments. The framework seeks to address real-time healthcare demands through dynamic offloading strategies, incorporating digital twins (DT) and social health determinants to personalise and improve healthcare interventions. Employing both partial and binary offloading strategies, multi-protocol communications are supported by the framework, ensuring seamless data exchange. The integration of DT and social health determinants into offloading decisions stands at the core of the methodology, rigorously tested in real-time settings. Iterative testing confirms the framework’s effectiveness, demonstrating a 10% enhancement in energy efficiency and a 20% reduction in network latency with 20 MEC nodes. The inclusion of 30 MEC nodes further reduced latency by 33.4% and power usage by 53.8% for data sizes up to 100 MB, evidencing significant advancements in healthcare technology integration. A significant gap in existing literature is bridged, and a new trajectory for technological innovation in healthcare systems is set by the research. The study underscores the potential of sophisticated offloading techniques to revolutionise healthcare delivery, offering a holistic solution to the challenges of data and computational management in medical contexts.

Keywords

• Adaptive cybersecurity task offloading (ACTO)
• digital twins healthcare
• energy efficiency in healthcare systems
• predictive healthcare interventions
• social health determinants