Smart spectrophotometric approach for the simultaneous determination of anti-asthma drugs in inhalation spray using least squares support vector machine and net analyte signal

Abstract

In this study, easy, fast, inexpensive, and accurate spectrophotometric technique with the help of least square support vector machine (LS-SVM) and net analyte signal (NAS) as chemometrics approaches were developed for the simultaneous determination of salmeterol (SMT) and fluticasone (FLU) in mixtures and anti-asthma inhalation spray. The related parameters of the LS-SVM model (sigma and gamma) were optimized and root mean square error (RMSE) were obtained 0.3075 and 0.4553 for SMT and FLU, respectively. Limit of detection (LOD) and limit of quantification (LOQ) dedicated to the calibration plot of the NAS method were 1.434, 1.089 μg mL−1 and 4.584, 3.399 μg mL−1 for SMT and FLU, respectively. The optimum number of components was evaluated by the minimum predicted residual error sum of squares (PRESS) value, which achieved 1.7738 and 1.0636 for SMT and FLU, respectively. Also, the mean recovery value of test set related to the 25 binary mixtures was 98.45% and 98.98% for SMT and FLU, respectively. Analysis of variance (ANOVA) test represented no significant difference significant between high-performance liquid chromatography (HPLC) (reference technique) and chemometrics methods. The LS-SVM and NAS can be a good alternative to expensive methods such as chromatography in the quality control section of different industries.