Artificial Intelligence-Integrated Biosensors for Antimicrobial Resistance Detection and Surveillance: A Review and Future Perspectives for Global Biosecurity

The review highlights that traditional methods for detecting antimicrobial resistance (AMR) — relying on culture and lab diagnostics — are too slow, expensive and often inaccessible, limiting effective surveillance worldwide. Instead, it argues that new biosensor technologies (electrochemical, optical, piezoelectric, paper-based and nanomaterial-based) combined with artificial intelligence (AI) and machine-learning can revolutionise AMR detection by enabling rapid, decentralized, real-time monitoring across clinical, veterinary and environmental settings under a unified One Health framework. These hybrid systems promise scalable, sensitive detection of resistance genes and pathogens, and could feed data into IoT-enabled networks for global AMR tracking. Nevertheless, the review also flags challenges: ensuring sensor reliability, addressing data scarcity, guaranteeing algorithm transparency and tackling regulatory, ethical and equity issues. The authors call for interdisciplinary collaboration, open-data standards, and investments to translate these innovations from proof-of-concept into real-world tools — potentially transforming global AMR surveillance and response.