Rapid identification research on Angelica sinensis from different producing areas based on electronic nose technology
GONG Jianting1,2 ZOU Huiqin3 WANG Jiayu4 WU Haozhong3 WANG Daqian1,2 LI Jiahui3 LIU Changli5 LI Li1,2
1.Beijing Institute of Chinese Medicine，Beijing 100035, China;
2.Beijing Chinese Medicine Hospital, Capital Medical University, Beijing 100010, China;
3.School of Chinese Materia Medica, Beijing University of Chinese Medicine, Beijing 102488, China;
4.Changchun Medical College, Jilin Province, Changchun 130031, China;
5.School of Traditional Chinese Medicine, Capital Medical University, Beijing 100069, China
Abstract：Objective To establish a fast, accurate and reliable method for identification of Angelica sinensis of different origin, based on the odor fingerprint of Angelica sinensis of different origin, and the discrimination model was established based on the difference odor. Methods Odor analysis of Angelica sinensis samples from different habitats was carried out by electronic nose. The maximum, average, maximum slope and integral values of sensor signals were extracted as characteristic parameters. Principal component analysis and support vector machine were used to analyze the characteristic parameters. The discriminant model was established. And examine the identification effect of combination characteristics on Angelica sinensis from different habitats. Results Principal component analysis could not distinguish the origin of Angelica sinensis. Support vector machine could quickly and accurately identify the habitat of Angelica sinensis. When a single feature was used as a feature parameter, the identification effect was general, and the average positive detection rate on the test set was 86.03%. When combined features were adopted as the feature parameters, the effect was better. The average positive detection rate of the combination of two features and three features was 86.76% and 89.71% respectively, and the positive detection rate of the combination of four features was the highest, which was 91.18%. Conclusion Electronic nose technology can accurately identify Angelica sinensis samples from different regions, providing new technologies and methods for rapid identification of the origin of traditional Chinese medicine. The selection and optimization of different pattern recognition algorithms and characteristic parameters provide more ideas for the application of electronic nose data mining in traditional Chinese medicine research.
拱健婷1,2 邹慧琴3 王佳宇4 吴浩忠3 王大仟1,2 李佳慧3 刘长利5 李莉1. 基于电子鼻技术的不同产地当归快速鉴别研究[J]. 中国医药导报, 2019, 16(28): 39-43.
GONG Jianting1,2 ZOU Huiqin3 WANG Jiayu4 WU Haozhong3 WANG Daqian1,2 LI Jiahui3 LIU Changli5 LI Li1,2. Rapid identification research on Angelica sinensis from different producing areas based on electronic nose technology. 中国医药导报, 2019, 16(28): 39-43.
2019, Vol. 16 
京公网安备 11010502046598号