Early detection and classification of fungal infection in garlic (A. sativum) using electronic nose

Abstract Garlic (Allium sativum) is an important crop with numerous benefits and uses. This plant is highly exposed to various pathogenic factors, including fungi. Fungi are the most distinct group responsible for planting diseases the loss involved. In this respect, detection of fungal pathogens infection in the early stages is a major challenge in food security to minimize losses as much as possible. Aroma investigation to detect fungal pathogens infection has been widely welcomed in this regard. In the present study, an electronic nose (E-nose) is utilized as a non-destructive and fast method for early detection of fungal infection on garlic that was synthetically infected with Fusariumoxysporum f. sp. Cepae (FU), Alternariaembellisia (syn. Embellisiaallii) (AL), and Botrytis allii (BO). Statistical analyses including ANOVA, PCA, LDA, SVM, and BPNN were employed to evaluate the aroma profile obtained by the E-nose. According to the obtained results, degradation occurs more quickly in the presence of infection. Due to the different destructive effects of each type of infection, the response changes of each sensor toward the aroma of various infection treatments were not the same. Thus, the E-nose can be used as a practical and beneficial tool to detect fungal infection on garlic in the early stages.