An ultra-sensitive, intelligent platform for food safety monitoring: Label-free detection of illegal additives using self-assembled SERS substrates and machine learning

To overcome the limitations of SERS in food safety monitoring, particularly significant interference from citrate ions, this study introduces an intelligent SERS-based platform for food safety monitoring. The platform utilizes sodium borohydride to activate silver nanoparticles, and calcium ions can facilitate the nanoparticles aggregation to promote self-assembly and the form of “hotspots”, but will also amplify citrate ions signal. Iodine ions was introduced to eliminate the interference of citrate signals and background fluorescence interference. The substrate achieved limit of detection of 100 fg/mL. Moreover, the innovative of spectral set “SERSome” enables comprehensive molecular fingerprint recognition, significantly enhancing accuracy. Furthermore, combined with machine learning enhances applicability for rapid and precise detection, and classification in food samples, and successfully applied to the monitoring of illegal additives in food. In summary, this system presents an intelligent, innovative detection platform for food safety, contributing to early prevention of foodborne illnesses.