Artificial intelligence algorithms-assisted biosensors in the detection of foodborne pathogenic bacteria: Recent advances and future trends

Background: Detecting foodborne pathogens is a critical aspect of food safety, requiring rapid, accurate, and reliable detection methods. Although conventional biosensing technologies have made significant progress, they still face challenges in sensitivity, accuracy, and adaptability, particularly in complex food matrices. The integration of artificial intelligence (AI) algorithms, particularly machine learning (ML) and deep learning (DL) techniques, shows great promise in overcoming these challenges and significantly enhancing biosensor performance. Scope and approach: This review examines the application of AI algorithms, focusing on ML and DL techniques, to enhance biosensors for detecting foodborne pathogens. It presents a comparative analysis of different AI models and recommends using algorithms tailored to various biosensor types, including surface-enhanced Raman spectroscopy (SERS), fluorescence, colorimetric, and electrochemical biosensors. The paper also explores the practical applications and limitations of these algorithms in food safety and outlines potential future directions. Key findings and conclusions: AI algorithms-assisted biosensors have significantly improved pathogen detection accuracy and efficiency. These algorithms allow biosensors to process complex multidimensional data in realtime, improving their ability to detect pathogens in diverse and challenging food samples. Despite notable advancements, challenges persist in algorithm adaptation and device compatibility. This review emphasizes the transformative potential of AI-assisted biosensors in advancing food safety detection technologies, focusing on driving future innovations and applications in the food industry.