Advancing Pandemic Preparedness in Healthcare 5.0: A Survey of Federated Learning Applications

Table 5

Comparison between FL and traditional pandemic preparedness methods.

Advantages	Strengths	Limitations
Data privacy protection	FL ensures privacy by not sharing raw patient data	Conventional methods often require sharing sensitive healthcare data
	Only aggregated model updates are exchanged among institutions	Patient data privacy concerns can pose ethical and legal challenges
Decentralized	FL empowers local decision-making, adapting to localized outbreaks	Traditional centralized systems may lead to slow and less adaptive decisions
Decision-making	Institutions or regions can make data-driven decisions based on local insights	Delays in data processing can hinder critical decision-making during a pandemic
Real-time surveillance	FL allows real-time insights as models are updated collaboratively	Data aggregation and analysis in traditional methods can be time-consuming
	Institutions can continuously update models, providing up-to-the-minute data	Delays in data processing can hinder critical decision-making during a pandemic
Resource-constrained environments	FL accommodates resource-constrained environments	Traditional methods may require substantial infrastructure and data centres
	Enables collaboration among institutions with varying resources	Institutions with limited resources may face barriers to participation
Research collaboration	FL fosters global research collaboration by allowing data analysis without data sharing	Traditional research often involves data silos and barriers to data sharing
	Researchers worldwide can collaborate on diverse datasets without compromising privacy	Data sharing limitations can hinder international research efforts