Data fusion of electronic noses and electronic tongues aids in botanical origin identification on imbalanced Codonopsis Radix samples

Abstract

Codonopsis Radix (CR) is an edible food and traditional Chinese herb medicine in China. Various varieties of Codonopsis Radix have different tastes. To make the flavor of processed food stable, two kinds of electronic sensory devices, electronic nose and electronic tongue, were used to establish a discrimination model to identify the botanical origin of each sample. The optimal model built on the 88 batches of samples was selected from the models trained with all combination of two pretreatment methods and three classification methods. A comparison were performed on the models trained on the data collected by electronic nose and electronic tongue. The results showed that the model trained on the fused dataset outperformed the models trained separately on the electronic nose data and electronic tongue data. The two preprocessing approaches could improve the prediction performance of all classification methods. Classification and Regression Tree approach performed better than Partial Least Square Discriminant Analysis and Linear Discriminant Analysis in terms of accuracy. But Classification and Regression Tree tends to assign the samples of minority class to the majority class. Meanwhile, Partial Least Square Discriminant Analysis keeps a good balance between the identification requirements of all the two groups of samples. Taking all the results above, the model built using the Partial Least Square Discriminant Analysis method on the fused data after z-score was used to identify the botanical origin of Codonopsis Radix.

Quality Assessment and Classification of Codonopsis Radix Based on Fingerprints and Chemometrics

“…The aforementioned findings are in line with what we anticipated, which was that the multifactorial parameters considerably changed how similar the various samples were to one another. Our findings concur with those from earlier studies [ 18 , 19 , 20 ]. It is well-known that the composition and content of chemical components of Chinese herbs change with soil conditions, cultivation measures and climate factors in the growth process [ 21 , 22 , 23 , 24 , 25 ].…”

“…This indicated that the quality of CR in Gansu and Shanxi Province was the best, followed by that in Sichuan Province, while that in Guizhou Province was relatively poor. These results demonstrated a large difference in the accumulations of 13 marker compounds (peaks 13,5,6,10,26,21,16,20,23,22,14,19,8) in different origins. We also analyzed the 16 marker compounds (peaks 22,27,19,7,12,21,5,26,9,4,1,28,8,6,24,17) in CR from different storage time and kneading processing (Figure 9B).…”

“…These results demonstrated a large difference in the accumulations of 13 marker compounds (peaks 13,5,6,10,26,21,16,20,23,22,14,19,8) in different origins. We also analyzed the 16 marker compounds (peaks 22,27,19,7,12,21,5,26,9,4,1,28,8,6,24,17) in CR from different storage time and kneading processing (Figure 9B). The contents of the 16 marker compounds were extremely high in K0Y samples, indicating that CR was of good quality after kneading.…”

Application of Data Fusion in Traditional Chinese Medicine: A Review

“…Finally, the sensor data were fused with the chromatography data, and the predictive model for the content constituents was established using PLSR. Wang et al [ 121 ] used two sensor devices, an e-nose, and an e-tongue, to establish a discriminative model of Codonopsis Radix, selecting the optimal model from all combinations of models trained by the two pre-processing methods and the three classification methods, and the results showed that the PLS-DA could well discriminate the original plant source of Codonopsis Radix after modeling. Miao et al [ 122 ] proposed a method for the fusion of e-nose and NIR data combined with SVM used for the identification of different species of ginsengs.…”

Identification of Codonopsis Radix from different sources by integrating high‐performance liquid chromatography fingerprinting with chemometric approaches and Lobetyolin determination

“…However, the chemical composition of these three varieties differs, resulting in varying quality and efficacy [9][10][11]. Currently, electronic nose and tongue technology, as well as chemical composition fingerprint technology have been frequently utilized by researchers to identify three different varieties of CR, which are effective identification methods [12][13][14][15][16]. However, these methods are complicated and require expensive equipment.…”