Generally, E-nose mimics human olfactory sense to detect and distinguish an odor or gasses or volatile organic compound from a few objects such as food, chemicals, explosive etc. Thus, E-nose can be used to measure gas emitted from food due to its ability to measure gas and odor. Principally, the E-nose operates by using a number of sensors to response to the odorant molecules (aroma). Each sensor will respond to their specific gas respectively. These sensors are a major part of the electronic nose to detect gas or odor contained in a volatile component. Information about the gas detected by sensors will be recorded and transmitted to the signal processing unit to perform the analysis of volatile organic compound (VOC) pattern and stored in the database classification, in order to determine the type of odor. Classification is a way to distinguish a mixture odor/aroma obtained from gas sensors in an electric signal form. In this paper, we discussed briefly about electronic nose, it’s principle of work and classification method and in order to classify food freshness.

Background: Several hematological indices have been already proposed to discriminate between iron deficiency anemia (IDA) and β‐thalassemia trait (βTT). The aim of the present study was to compare the diagnostic performance of different hematological discrimination indices with statistical methods such as decision trees to discriminate IDA from βTT.Methods: Consisting of 1178 patients with hypochromic microcytic anemia (708 patients with βTT and 470 patients with IDA), this cross-sectional study intended to compare the diagnostic performance of 43 hematological discrimination indices and tree-based methods such as J48, CART, Evtree, Ctree, QUEST, CRUISE as well as GUIDE to discriminate IDA from βTT. Moreover, multidimensional scaling and cluster analysis were used to identify the homogeneous subgroups of discrimination methods with similar performances.Results: All the classification tree algorithms showed acceptable accuracy measures for discrimination between IDA and βTT in comparison with other hematological discrimination indices. The results indicated that CRUISE tree algorithm had better diagnostic performance and efficiency among other discrimination methods. In turn, this tree algorithm showed the high Youden's index (88.03%), accuracy (94.57%), diagnostic odds ratio (311.63) and F-measure (95.54%) in the differential diagnosis of IDA from βTT. In addition, AUC of this algorithm indicated more precise classification with a value of 0.94 and this model was found to have excellent diagnostic accuracy. Also, CRUISE tree algorithm showed that Mean corpuscular volume can be considered as the main variable in discrimination as it extracted six homogenous subgroups of patients.Conclusions: CRUISE tree algorithm as a powerful method in data mining techniques can be used to develop accurate differential methods along with other laboratory parameters to discriminate IDA from βTT.