A Comparative Study of optimized conditions of QuEChERS to determine the pesticide multiresidues in Lycium barbarum using Response Surface Methodology and Genetic Algorithm-Artificial Neural Network

The present study sought to improve the QuEChERS (quick, easy, cheap, effective, rugged, safe) method to determine the pesticide residues in Lycium barbarum. The QuEChERS conditions were optimized by coupling an artificial neural network (ANN) technique with the genetic algorithm (GA) method, and the response surface methodology (RSM). The results revealed that GA-ANN was more accurate than RSM. Meanwhile, the limit of detection and the limit of quantitation of the QuEChERS method which was optimized by GA-ANN were 0.023–0. 0644 mg·kg−1 and 0.0077–0. 2149 mg·kg−1, respectively. The recoveries of the three levels were 92.36–108. 79% with the Relative Standard Deviation ranging from 0.43% to 6. 69%. The regression coefficient for all calibration curves were greater than 0.9960. Overall, the improved QuEChERS method exhibited high accuracy and good recovery toward determining the pesticide residues in Lycium barbarum.