Adulteration identification strategy for Acanthopanax Senticosus based on data fusion of portable mass spectrometry and near-infrared spectroscopy

Abstract

Acanthopanax senticosus is extensively studied worldwide. Due to high price, illegal activities, particularly adulteration, have proliferated. To identify adulterated Acanthopanax senticosus, it is necessary to target both the high content of quality-related compounds and the overall physical and chemical properties. This study utilized data fusion methods to integrate portable Near-Infrared (NIR) spectroscopy and Portable Mass spectrometry (PMS) technologies for the identification of adulteration in Acanthopanax senticosus. A high-level data fusion for adulteration detection was achieved through fuzzy algorithms. The data fusion model exhibited excellent performance on the prediction set, achieving an accuracy of 0.96. The regression model, constructed using a weighted average algorithm for the quantification of adulteration percentages, demonstrated predictive capability with R²_P of 0.9664, RMSEP of 0.0535 g/g. The advanced fusion models outperformed single-technology models, indicating that the combination of PMS and NIR with data fusion strategies is practical for both identification and quantification of adulteration in Acanthopanax senticosus.