Machine learning-driven fluorescent sensor array using aqueous CsPbBr3 perovskite quantum dots for rapid detection and sterilization of foodborne pathogens

Abstract

With the growing global concern over food safety, the rapid detection and disinfection of foodborne pathogens have become critical in public health. This study presents a novel machine learning-driven fluorescent sensor array utilizing aqueous CsPbBr₃ perovskite quantum dots (PQDs) for the rapid identification and eradication of foodborne pathogens. The relative signal intensity changes (ΔRGB) generated by the sensor array were analyzed using the machine learning algorithm-Support Vector Machine (SVM). The study achieved the identification and recognition of five pathogens and their mixtures within a concentration range of 1.0 × 10³ to 1.0 × 10⁷ CFU/mL with an accuracy rate of 100 %, and the limits of detection (LOD) for the pathogens were found to be low. Additionally, the array also showed excellent performance in the identification of pathogens in tap water, achieving an accuracy rate of 100 %. Furthermore, the fluorescent sensor array was capable of inactivating the pathogens with an efficiency of over 99 % within 30 min post-detection. This development provides an efficient and reliable tool for the field of food safety detection.

Keywords: Fluorescent sensor array; Foodborne pathogens; Machine learning; Perovskite quantum dots; Rapid detection.