The Combination of Machine Learning Tools with the Rapid Visco Analyser (RVA) to Enhance the Analysis of Starchy Food Ingredients and Products

Abstract

This review discusses how the integration of machine learning (ML) tools enhances the analytical capabilities of the Rapid Visco Analyser (RVA), aiming to provide a deeper understanding of the starch gelatinization in different starchy food ingredients and products. The review also discusses some of the limitations of RVA as a tool for assessing the pasting and viscosity behavior of starch, emphasizing the potential of different ML tools such as principal component analysis (PCA) and partial least squares (PLS) regression to offer a better analytical approach. Examples of the utilization of ML combined with RVA to enhance the analysis of starch and non-starch ingredients are also provided. Furthermore, the importance of preprocessing techniques, such as derivatives, to improve the quality and interpretability of RVA profiles is discussed. The aim of this review is to provide examples of the utilization of RVA combined with ML tools in starchy food ingredients and products.

Keywords

• rheology
• RVA
• starch
• viscosity
• breeding
• cereals