Microwave-assisted aqueous extraction of bioactive components from Malabar spinach (Basella alba) leaves and its optimization using ANN-GA and RSM methodology

The paper discusses the effect of microwave assisted aqueous extraction conditions on the recovery and quantification of bioactive compounds of Basella alba leaves. The experiment was carried out using the Box-Behnken design of response surface methodology (RSM) with microwave power (100-300 W), treatment time (5-15 min), and feed/solvent ratio (0.025-0.05 w/v) as independent variables. The response variables, total phenolic content (TPC), total flavonoid content (TFC), and antioxidant activity (AA) were predicted individually using both RSM and artificial neural network-genetic algorithm (ANN-GA). Experimental values of TPC, TFC and AA (% of DPPH scavenging assay) ranged from 2.64 to 5.46 mg GAE g-1, 7.38 to 15.71 mg QE g(-1) and 0.12 to 0.32 mg GAEAC g(-1) respectively. The predicted values of TPC, TFC, and AA for the optimized conditions extracted using RSM are 6.21 mg GAE g(-1), 14.29 mg QE g(-1), and 0.25 mg GAEAC g(-1), respectively, whereas using ANN-GA were 6.23 mg GAE g(-1), 11.2 mg QE g(-1), and 0.24 mg GAEAC g(-1), respectively. When compared to RSM, ANN-GA demonstrated a greater value of R-2 and lower values of other statistical parameters. Additionally, the predicted value of ANN-GA was more closely aligned with the experimental value. Therefore, ANN-GA can be considered the best model for the optimization and modeling of aqueous MAE of bioactive components from Basella alba leaves.