Nano-effect multivariate fusion spectroscopy combined with chemometrics for accurate identification the cultivation methods and growth years of Dendrobium huoshanense

Abstract

Dendrobium huoshanense (DHS) is an expensive Chinese herb with medicinal and edible functions. The quality of DHS is closely related to cultivation methods and years. In this work, nano-effect near and mid infrared spectra of 320 DHS samples with four cultivation methods and four years were obtained, and the feature vectors were obtained using a mid-level data fusion strategy with significant variable extraction and integration using variable importance for the projection (VIP). The feature vectors then combined with partial least squares discriminant analysis (PLS-DA) and orthogonal partial least squares discriminant analysis (OPLS-DA) to identify the cultivation years and methods. The result showed that it is difficult to discriminate effectively by a single spectrum, and the accuracy of PLS-DA discrimination based on raw or simple fusion of raw spectra is only 95.5% and 84.4% or 99.6% and 88.5%, while the accuracy of discrimination based on nano-effect feature fusion spectra can be improved to 100%, and PLSR was used to quantify the cultivation year of the DHS with good linearity and accuracy. The nano-effect spectrum is based on the reaction of tetraphenyl zinc porphyrin (ZnTPP) with polysaccharides, flavonoids and alkaloids in DHS, which amplifies the spectral difference. This study demonstrates that multi-spectral feature vectors fusion method could potentially be a reliable analytical method to distinguish the cultivation methods and years in DHS herbs.