Use of electronic nose and trained sensory panel in the evaluation of tomato quality

In this paper the performances of an electronic nose based on metalloporphy rin-coated quartz microbalance sensors and of an experienced panel of seven human assessors in the evaluation of gases derived from degradation reacti ons in tomatoes are presented and discussed. The performances are measured in terms of the capability of both systems to distinguish between samples o f different quality coming from conventional and organic production systems . The study deals with the application of pattern recognition techniques ba sed on either multivariate statistical methods (PCA, GPA) or artificial neu ral networks using a self-organising map (SOM). The response pattern of the sensor array and the sensory data are analysed and compared using these me thods. Similarities in the classification of the data by electronic nose an d human sensory profiling are found. (C) 2000 Society of Chemical Industry.