Quality Traits of Sourdough Bread Obtained by Novel Digital Technologies and Machine Learning Modelling

There is no author summary for this article yet. Authors can add summaries to their articles on ScienceOpen to make them more accessible to a non-specialist audience.

Abstract

Sourdough bread (SB) has increased popularity due to health benefits and higher interest in artisan breadmaking due to social isolation during the COVID-19 pandemic. However, quality traits and consumer assessment are still limited to complex laboratory analysis and sensory trials. In this research, new and emerging digital technologies were tested to assess quality traits of SB made from six different flour sources. The results showed that machine learning (ML) models developed to classify the type of wheat used for flours (targets) from near-infrared (NIR) spectroscopy data (Model 1) and a low-cost electronic nose (Model 2) as inputs rendered highly accurate and precise models (96.3% and 99.4%, respectively). Furthermore, ML regression models based on the same inputs for NIR (Model 3) and e-nose (Model 4) were developed to automatically assess 16 volatile aromatic compounds (targets) using GC-MS as ground-truth. To reiterate, models with high accuracy and performance were obtained with correlation (R), determination coefficients (R2), and slope (b) of R = 0.97; R2 = 0.94 and b = 0.99 for Model 3 and R = 0.99; R2 = 0.99 and b = 0.99 for Model 4. The development of low-cost instrumentation and sensors could make possible the accessibility of hardware and software to the industry and artisan breadmakers to assess quality traits and consistency of SB.

Related collections

Most cited references 52

Record: found
Abstract: not found
Article: not found

Sourdough volatile compounds and their contribution to bread: A review

Bernard Onno, Carole Prost, Cécile Pétel (2017)

0 comments Cited 53 times – based on 0 reviews      Review now

Bookmark

Record: found
Abstract: found
Article: not found

Lifestyles of sourdough lactobacilli – Do they matter for microbial ecology and bread quality?

Michael G Gänzle, Jinshui Zheng (2018)

Sourdough is used in production of (steamed) bread as leavening agent (type I sourdoughs) or as baking improver to enhance flavour, texture, and shelf life of bread (type II sourdoughs). The long-term propagation of sourdoughs eliminates dispersal limitation and consistently leads to sourdough microbiota that are composed of host adapted lactobacilli. In contrast, community assembly in spontaneous cereal fermentations is limited by dispersal and nomadic or environmental lactic acid bacteria are the first colonizers of these sourdoughs. Propagation of sourdoughs for use as sole leavening agent (type I sourdoughs) dictates fermentation conditions that select for rapid growth. Type I wheat- and rye sourdoughs are consistently populated by insect-adapted lactobacilli, particularly Lactobacillus sanfranciscensis, which is characterized by a small genome size and a restricted metabolic potential. The diverse fermentation conditions employed in industrial or artisanal Type II sourdough fermentation processes also result in a more diverse microbiota. Nevertheless, type II sourdoughs are typically populated by vertebrate host adapted lactobacilli of the L. delbrueckii and L. reuteri groups. Metabolic traits of host-adapted lactobacilli that enhance competitiveness in intestinal ecosystems also provide technological functionality in bread making. Examples include formation of exopolysaccharides, arginine-, glutamine- and glutamate based mechanisms of acid resistance, and glycosyl hydrolases that reduce FODMAP levels in sourdough and sourdough bread. In conclusion, consideration of the lifestyle of sourdough lactobacilli facilitates the selection of competitive and functional sourdough starter cultures.