Ultrasound-mediated hydration of finger millet: Effects on antinutrients, techno-functional and bioactive properties, with evaluation of ANN-PSO and RSM optimization methods

Abstract

Finger millet, rich in nutrients, faces bioavailability limitations due to antinutrients like phytates and tannins that can be reduced by ultrasound mediated hydration (USH). Here, USH process of finger millet was optimized by varying ultrasound amplitude, water to grain ratio (W:G), treatment time, and frequency for reducing antinutrients and improving techno-functional attributes. USH resulted in a maximum reduction of 73% and 71% in phytates and tannins, respectively. The process was modeled using artificial neural network (ANN) and response surface methodology (RSM). ANN outperformed RSM in process prediction, and particle swarm optimization (ANN-PSO) suggested optimal conditions: 76% amplitude, W:G of 3.5:1, 17.5 min treatment time at 40 kHz. USH samples showed higher β-sheet, β-turn, and random coil proportions, with lower α-helix levels. Multivariate analysis also identified higher amplitude and frequency, with shorter treatment time as desirable USH conditions. USH could aid in enhancing commercial viability and nutritional quality of finger millet.