A comparative study of RSM and ANN models for predicting spray drying conditions for encapsulation of Lactobacillus casei

Abstract

Background and Objectives

The aim of this study was to develop a wall material using pea protein isolate and pectin to optimize the encapsulation of Lactobacillus casei by spray drying. Response surface methodology (RSM) and artificial neural network (ANN) were used to analyze the effect of processing parameters.

Findings

The results showed that both RSM and ANN could be used to successfully characterize the experimental data, although ANN demonstrated greater predictive accuracy than RSM due to a higher R² and lower mean square error (MSE).

Conclusion

ANN was observed to show more suitability than RSM. The encapsulation efficiency (90.7%), yield (45.5%), and wettability (169 s) of spray-dried probiotic powder obtained under optimal spray drying conditions (inlet air temperature (132°C); feed flow rate (9.5 mL/min) and pea protein isolate concentration (7.1%)) were observed to be not significantly different (p < .05) from predicted values for all three parameters, demonstrating the validity of applied model.

Significance and Novelty

In this study, production technology of vegan base probiotic powder has been developed using mathematical modeling through the spray-drying method. Therefore, this data can be useful for food processing industries to develop a high-quality probiotic powder through spray drying.