Machine learning driven metal oxide-based portable sensor array for on-site detection and discrimination of mycotoxins in corn sample

Cereals, grains, and feedstuffs are prone to contamination by fungi during various stages from growth to storage. These fungi may produce harmful mycotoxins impacting food quality and safety. Thus, the development of quick and reliable methods for on-site application is crucial for ensuring food safety and quality monitoring. Herein, we have developed an efficient sensor array based on hierarchically modified metal oxides with azodye-based metal complexes for on-site detection and segregation of harmful mycotoxins present in corn samples. The functionalized material has been fully characterized utilizing various sophisticated techniques. The sensor array successfully detected and differentiated five different mycotoxins with 100 % efficiency, validated by linear discriminant analysis (LDA) score plots. The limit of detection, as determined from calibration curves, ranges from 0.02 to 0.09 ppm for the respective mycotoxins. Additionally, the sensor array has also demonstrated 100 % accuracy in discriminating binary and ternary ratios of mycotoxins in real sample analyses.