基于气相色谱-离子迁移谱及机器学习的不同产地香附饮片的快速鉴别OA

Gas chromatography-ion mobility spectrometry(GC-IMS)was employed to analyze the volatile components of Cyperi Rhizoma from different origins.Partial least squares-discriminant anal‑ysis(PLS-DA)was applied to screen differential characteristic components.Based on these charac‑teristic components,9 machine learning algorithms such as SVM-L were constructed to develop rap‑id discrimination models for Cyperi Rhizoma from different origins.Fifty-nine volatile compounds were identified in Cyperi Rhizoma decoction pieces from five provinces,including alcohols,esters,aldehydes,ketones,and unsaturated hydrocarbons.Ten characteristic components were screened by PLS-DA analysis,such as 1,4-dimethylbenzene,3-methyl-1-pentanol,and ethyl acetoace‑tate.Furthermore,nine machine learning models all demonstrated excellent predictive performance(Accuracy=1),indicating good potential for practical application.This study provides a simple and rapid method for discrimination and identification of Cyperi Rhizoma from different geographical ori‑gins,while also offering a reference for establishing a quality evaluation system for Cyperi Rhizoma.