Near infrared spectroscopy discrimination and pattern recognition of Chinese Pogostemon cablin and Agastache rugosa

Abstract

Pogostemon cablin (Blanco) Benth. and Agastache rugosa (Fisch. et Mey.) O. Ktze. are commonly used traditional Chinese medicines. They belong to the family Lamiaceae and have very similar macroscopical features, making it difficult to differentiate them. To establish a discriminatory method for P. cablin of various origins that has abundant in chemical components and has high medicinal value, we utilized near-infrared spectroscopy (NIRS) in this study to gather spectral information in the ranging from 4000 to 12,000 cm−1. Subsequently, we employed principal component and cluster analyses to develop qualitative discriminant models. The results showed that spectral pretreatment improved the classification and aggregation characteristics of P. cablin and A. rugosa in principal component and cluster analyses. The percentage of samples correctly classified using the principal component analysis in calibration and validation set was 99.9 ​%. This study results demonstrate that NIRS with pattern recognition can be utilized as a simple and rapid method for distinguishing between P. cablin and non-P. cablin, as well as A. rugosa.

Keywords

• Near-infrared spectroscopy
• Pogostemon cablin (Blanco) Benth
• Agastache rugosa (Fisch. et Mey.) O. Ktze.
• Principal component analysis
• Cluster analysis
• Pattern recognition