Development of innovative artificial neural networks for simultaneous determination of lapatinib and foretinib in human urine by micellar enhanced synchronous spectrofluorimetry.

A highly selective and simple micellar synchronous spectrofluorimetric method was described for simultaneous analysis of two tyrosine kinase inhibitors (TKIs); namely lapatinib (LPB) and foretinib (FTB) in human urine. The method depended on measuring synchronous fluorescence of the two drugs in micellar media composed of cremophor RH 40 (Cr RH 40) surfactant using feed-forward and cascade-forward neural networks preceded by genetic algorithm for data manipulation. Different experimental conditions that affect fluorescence of the cited drugs are optimized including pH, diluting solvent, surfactant's type and concentration. A training set of nine mixtures containing different concentrations of both drugs was prepared for models' construction. Extra validation set composed of other nine mixtures was prepared to validate prediction performance for the constructed models. Root mean square error of prediction (RMSEP) was used as a tool to compare prediction power of each model. The method was extended for quantification of LPB and FTB in spiked human urine.