Rheological Analyses and Artificial Neural Network as Optimization Tools to Predict the Sensory Perception of Cosmetic Emulsions

Abstract

Pharmaceutical, cosmetic and personal care products are mainly based on emulsions and their rheological behavior can be a critical factor for successful use. Thus, rheological analysis is a promising tool, since the stability, sensory aspects and processing parameters can be assessed. This work presents the rheological analyses of 39 samples of emulsions and the use of data obtained in a tool based on artificial neural networks (ANN), in order to predict the sensory performance of cosmetic emulsions. The storage (G') and loss (G") moduli, yield stress and thixotropy were measured experimentally and used in the ANN model. The correlation of the results obtained in the simulations with sensory tests performed with consumers showed accuracy of 60-84%. The reported results demonstrate that the prediction of sensory perception based on rheological analyses offers a very useful strategy for further studies and can support the development of new products in less time.

Textural analyses and artificial neural network as optimization tools to predict the sensory perception of cosmetic emulsions

“…The simulations carried out in this study show that the texture results are effective to be used in the sensory perception predicting. The comparison of the results obtained in this study with those observed in the predictions performed with rheological data, described previously by authors (Franzol et al, 2021), shows that the textural data results in simulations with better performance. In addition, the predictions based on textural data showed adherence with more attributes (7) and predictions based on rheological results with only 5 attributes.…”

“…As already mentioned in the previous work of authors (Franzol et al, 2021), which involved the ANN modeling inputted with rheological data, the accuracy of predictive models are based on the space generated by the values of their corresponding variables. Thus, predictions generated with points outside this space do not have error estimates.…”

“…To preserve the characteristics of each product, without exposing details of the formulations and respective manufacturers, the samples were randomly assigned a new code (Table 2). The same sampling was used in a previous work by the authors to predict the sensory perception, but using rheological measurements and ANN model, with good results (Franzol, Banin, Brazil, & Rezende, 2021).…”

“…If very much approved the product was classified as 5 and if very poor perception, it was classified as 1. This procedure was based on the literature (Franzol et al, 2021). Then, the collected data were organized and analyzed using two methods adopted in the market research: the Just About Right (JAR) and Top Two Box (TTB) methods, in accordance with to the Society of Sensory Professionals and the literature (Franzol et al, 2021; SSP, 2022; Vieira, 2019).…”

“…The statistical model used in this study was described in a previous work of the authors (Franzol et al, 2021) based on the artificial neural networks, with one hidden layer, according to the literature (Hayes & Preacher, 2014; Haykin, 1998; Pimentel, 2005; Silva, 2016), using internal validation and 25% of the population for this. Haykin (Haykin, 1998) suggests that the data set can be shared in 50% of the data for training, 25% for validation, and 25% for testing.…”