Statistical and neural intelligence modeling for basil seed mucilage extraction optimization: Implications for sustainable and cost-effective industrial plant products

Abstract

The upsurge in exigency of environmental-friendly and vigorous plant-based products has spurred a substantial increase in the use of plant-based biopolymers, most conspicuously mucilage and gums. Plant extracted mucilage encompasses a group of complex macromolecules and is renowned for its stabilization, thickening and gelling properties, besides its drug delivery potential. Basil (Ocimum basilicum L.) seed mucilage embodies a polysaccharide of plant origin and is often characterized by its branched carbohydrate structure. Its consumption not only offers prospective health advantages but also aligns with an eco-friendly paradigm. In this study, the optimization of the extraction yield of basil seed mucilage (BSM) was done using response surface methodology and artificial neural network. The experimental design encompassed four parameters, namely pH, temperature, contact time and seed/water ratio, using a 3-level central composite design. The response surface methodology (RSM) and genetic algorithm feedforward neural network (ANN) were employed to predict and evaluate the optimal extraction conditions. The optimal conditions for the extraction yield of BSM were determined to be 7 pH, 56 °C temperature, 6 h of contact time and a 1:30 (w/v%) seed/water ratio. These conditions resulted in a BSM extraction actual yield of 9.94%, which was close to the RSM and ANN predicted values, demonstrating the effectiveness of this approach for optimizing the plant-based polymer extraction process parameters.