Early detection and classification of pathogenic fungal disease in post-harvest strawberry fruit by electronic nose and gas chromatography-mass spectrometry

Strawberry fruit decay caused by fungal infection usually results in considerable losses during post-harvest storage; thus, discerning the decay and infection type in the early stage is necessary and helpful for reducing the losses. In this study, three common pathogenic fungi belonging to Botrytis sp., Penicillium sp. and Rhizopus sp. were individually inoculated into ripe strawberry fruits; non-inoculated fruits were used as controls. The strawberry fruits were stored at 5 ± 1 °C for 10 days. During storage, inoculated fruits began rotting on day 2, while control fruits began rotting on day 4. The volatile compounds emitted by the fruits were analysed by an electronic nose (E-nose) and gas chromatography-mass spectrometry (GC-MS). Principal component analysis (PCA) showed a clear discrimination in decay on day 0, day 2 and day 4 and the infection type on day 2 after fungal inoculation based on 5 selected sensors of E-nose. The discrimination accuracy of the fungal infection type of strawberry fruits for the four groups reached 96.6% by using multilayer perceptron neural network model. GC-MS results of the four strawberry fruit groups on day 2 identified several key characteristic volatile compounds for each infection treatment, compared with the control. Therefore, E-nose was able to realise the early diagnosis of fungal disease, in addition to an accurate classification of the pathogenic fungal type in the fruits during post-harvest storage.