Augmented Visualization and Gas Chromatographic Fingerprinting Unveil Hidden Clues in Food Volatilome

Published on: 

Innovative gas chromatography (GC) study combines computer vision and chromatographic fingerprinting to uncover diagnostic signatures in food volatilome.

Cutting-edge technology is revolutionizing our ability to explore the chemical composition of food samples and unravel diagnostic signatures within the volatile compounds they release. A groundbreaking study published in the Journal of Chromatography A introduces an innovative approach that combines computer vision and chromatographic fingerprinting to analyze comprehensive two-dimensional gas chromatographic (2D-GC) patterns (1). Led by Chiara Cordero at the Università di Torino in Italy, this research aims to unlock valuable insights into the intricate world of the food volatilome and its relation to biological phenomena.

The food volatilome refers to the complex mixture of volatile compounds present in food. These volatile compounds are responsible for the aroma and flavor of food products. Analyzing the food volatilome allows scientists to understand the chemical composition and sensory characteristics of different foods. By studying the food volatilome, researchers can identify specific compounds that contribute to the aroma and flavor profiles of foods, which can be used for quality control, product development, and understanding the impact of processing and storage conditions on food quality.

Computer vision, an artificial intelligence technique, plays a pivotal role in this research by enabling computers to derive meaningful information from digital images. By harnessing this power, researchers can process vast amounts of data and make data-driven decisions. In the context of comprehensive two-dimensional chromatography, which provides detailed but unstructured information about a sample's chemical composition, computer vision acts as a key tool for raw data exploration and analysis.

The study introduces a novel workflow that incorporates pattern recognition algorithms, such as combined untargeted and targeted fingerprinting. One of the key features of this workflow is the generation of composite Class Images that represent chemical patterns within samples. These Class Images undergo effective re-alignment and registration against a comprehensive feature template, enabling augmented visualization through comparative visual analysis. By leveraging this innovative approach, researchers can capture the evolution of volatile components along the production chain and assess the impact of different microbial cultures on the final product's volatilome.


To demonstrate the power of their methodology, the researchers focused on an illustrative application: artisanal butter production. By analyzing a sample set ranging from raw sweet cream to ripened butter, they were able to gain insights into the volatile component dynamics and the influence of microbial cultures. This research showcases the advantages of the proposed workflow compared to traditional pairwise comparison methods, as it allows for the realignment and comparison of both targeted and untargeted chromatographic features belonging to Class Images. Moreover, the workflow accommodates the inherent biological variability observed in chemical patterns across multiple samples.

The integration of computer vision and chromatographic fingerprinting on comprehensive two-dimensional gas chromatographic patterns opens up new avenues for exploring the intricate world of the food volatilome. By unveiling hidden clues within complex chemical compositions, this research paves the way for enhanced quality control, optimization of food production processes, and a deeper understanding of the biological phenomena associated with specific chemical signatures. This groundbreaking research sets the stage for further advancements in the field, where computer vision and chromatographic fingerprinting converge to unravel the secrets hidden within the volatile compounds of our food.


(1) Caratti, A.; Squara, S.; Bicchi, C.; Tao, Q.; Geschwender, D.; Reichenbach, S. E.; Ferrero, F.; Borreani, G.; Cordero, C. Augmented visualization by computer vision and chromatographic fingerprinting on comprehensive two-dimensional gas chromatographic patterns: Unraveling diagnostic signatures in food volatilome. J. Chromatogr. A 2023, 1699, 464010.DOI: