Application of E-nose technology combined with artificial neural network to predict total bacterial count in milk

摘要

Total bacterial count (TBC) is a widely acceptedindex for assessing microbial quality of milk, and cultivation-based methods are commonly used as standardmethods for its measurement. However, these methodsare laborious and time-consuming. This study proposesa method combining E-nose technology and artificialneural network for rapid prediction of TBC in milk.The qualitative model generated an accuracy rate of100% when identifying milk samples with high, medium,or low levels of TBC, on both the testing andvalidating subsets. Predicted TBC values generated bythe quantitative model demonstrated strong coefficientof multiple determination (R~2 > 0.99) with referencevalues. Mean relative difference between predicted andreference values (mean ± standard deviation) of TBCwere 1.1 ± 1.7% and 0.4 ± 0.8% on the testing andvalidating subsets involving 24 and 28 tested samples,respectively. Paired t-test implied that the differencebetween predicted and reference values of TBC was insignificantfor both the testing and validating subsets.As low as ～1 log cfu/mL of TBC present in testedsamples were precisely predicted. Results of this studyindicated that combination of E-nose technology andartificial neural network generated reliable predictionsof TBC in milk. The method proposed in this study wasreliable, rapid, and cost efficient for assessing microbialquality milk, and thus would potentially have realisticapplication in dairy section.