Optimization of ultrasound-assisted extraction of bioactive chemicals from Hemidesmus indicus (L.) R.Br. using response surface methodology and adaptive neuro-fuzzy inference system

This study was designed to optimize the ultrasound-assisted extraction (UAE) of bioactive chemicals from Hemidesmus indicus (L.) R.Br. through RSM (response surface methodology) and ANFIS (adaptive neuro-fuzzy inference system). The effect of four independent parameters, methanol concentration (X-1: 55-65%), temperature (X-2: 30-40 degrees C), time (X-3: 15-20 min) and particle size (X-4: 0.5-1.00 mm) at five levels (- 2 ,- 1, 0, + 1, + 2) with respect to dependent parameters, total polyphenols content (TP) (y(1)), total flavonoids content (TF) (y(2)), %DPPHsc (y(3)), %ABTSsc (y(4)) and %H(2)O(2)sc (y(5)) were selected. The optimal extraction condition was observed at X-1 = 65%, X-2 = 40 degrees C, X-3 = 20 min and X-4 = 0.5 mm; under this circumstance, y(1) = 352.85 mg gallic acid equivalents (GA)/g, y(2) = 300.204 mg rutin equivalents (RU)/g and their antioxidant potentials (y(3) = 81.33%, y(4) = 65.04%, and y(5) = 71.01%) has been attained. ANFIS was used to compare and confirm the optimized extraction parameter values. Further, GC-MS and LC-MS were performed to investigate the bioactive chemicals present in the optimized extract.