Application of near infrared spectroscopy analysis method in rapid detection of sodium aescinate / 药学实践杂志

ABSTRACT

Objective To establish a fast detection method of sodium aescinate by using near infrared (NIR) spectroscopy analysis method for the determination of the content of sodium aescinate for injection. Methods OPUS software was used to optimize the collected spectrum. PLS algorithm and factorization algorithm were used to establish quantitative model and qualitative model. Results The correlation coefficient of the quantitative model reached 0.9926, the RMSECV deviation was 0.253. The deviation between the predicted value of the sample and the true measured value was less than 5%, which could accurately predict the content of sodium aescinate. Conclution The qualitative model can effectively distinguish the samples of other varieties that have not participated in the modeling, and provide a reference for the rapid screening of the drug.