Research on rapid quality identification method of Panax notoginseng powder based on artificial intelligence sensory technology and multi-source information fusion technology.

• A new method was developed for rapid quality evaluation of Panax notoginseng powder (PNP). • In identifying PNP grade and adulteration, the highest accuracy reached 100 %. • The evaluation system for adulterated PNP (AD-PNP) was established. • We found a relationship between the AIS information and the saponin content of AD-PNP. • The electronic tongue was more suitable for the quality evaluation of AD-PNP. Panax notoginseng powder (PNP) has high medicinal value and is widely used in the medical and health food industries. However, the adulteration of PNP in the market has dramatically reduced its efficacy. Therefore, this study intends to use artificial intelligence sensory (AIS) and multi-source information fusion (MIF) technology to try to establish a quality evaluation system for different grades of PNP and adulterated Panax notoginseng powder (AD-PNP). The highest accuracy rate reached 100% in identifying PNP grade and adulteration. In the prediction of adulteration ratio and total saponin content, the optimal determination coefficients of the test set were 0.9965 and 0.9948, respectively, and the root mean square errors were 0.0109 and 0.0123, respectively. Therefore, the grade identification method of PNP and the evaluation system of AD-PNP based on AIS and MIF technology can rapidly and accurately evaluate the quality of PNP. [ABSTRACT FROM AUTHOR]