Sensory-biased autoencoder enables prediction of texture perception from food rheology.

Journal: Food research international (Ottawa, Ont.)
PMID:

Abstract

Understanding how the physical properties of food affect sensory perception remains a critical challenge for food design. Here, we present an innovative machine learning strategy to decode the complex relationships between non-Newtonian rheological attributes of liquid foods and their perceived texture. A unique and key aspect of our approach is the implementation of an autoencoder neural network that incorporates sensory scores as a decoder bias during training. This enables the autoencoder to effectively identify non-linear, non-injective relationships between shear-thinning properties and perceived thickness, even when trained on a small dataset. This strategy offers a promising approach for advancing food product development by aiding the design of carefully tailored sensory experiences.

Authors

  • Paul M Kraessig
    Transport Phenomena Laboratory, Department of Food Science, Purdue University, West Lafayette, IN, USA.
  • Shyamvanshikumar P Singh
    Transport Phenomena Laboratory, Department of Food Science, Purdue University, West Lafayette, IN, USA.
  • Jiakai Lu
    Department of Food Science, University of Massachusetts, Chenoweth Laboratory, 01003 Amherst, MA, United States; Department of Mechanical and Industrial Engineering, University of Massachusetts, 01003 Amherst, MA, United States. Electronic address: jiakailu@umass.edu.
  • Carlos M Corvalan
    Department of Food Science, Purdue University, 47906, West Lafayette, Indiana, United States.