Researchers at McGill University have developed an innovative AI-powered method that verifies the origin of honey, ensuring that what is on the label matches what’s inside the jar. This groundbreaking study, published in Analytical Chemistry, presents a potential solution to a longstanding issue in the food industry.
Stéphane Bayen, the senior author and Associate Professor at McGill’s Department of Food Science and Agricultural Chemistry, stated, ‘Honey is one of the most fraud-prone commodities in global trade. It often involves mislabeling regarding its production location or the types of flowers from which bees collected nectar.’
Many consumers are willing to pay a premium for honey derived from a single flower variety, attracted by its unique flavors and health benefits. However, some producers either intentionally or unknowingly mislabel their honey, complicating efforts to track its true origins.
The new method enables researchers to determine the types of flowers visited by bees in the production of specific honey, aiding consumers in making informed purchasing decisions.
While traditional authentication methods rely on pollen analysis, which fails after the honey is processed or filtered, the new approach utilizes high-resolution mass spectrometry to create a unique chemical ‘fingerprint’ of honey. Machine learning algorithms then analyze this fingerprint to validate the honey’s origins.
‘Currently, identifying the true source of honey can take days. Our method can do this in minutes, even for processed honeys where standard techniques falter,’ Bayen explained.
Consumer and Producer Protection
The demand for locally sourced honey, such as Quebec’s blueberry honey, is increasing as consumers prioritize local purchasing. This new technique could protect both consumers and ethical beekeepers from dishonest practices.
‘People deserve to know that their honey is what it claims to be, and honest producers deserve protection from fraudulent competitors,’ Bayen asserted. The researchers hope that their method will be adopted by food inspection agencies globally and plan to explore its application in other food products vulnerable to mislabeling.
More information: Shawninder Chahal et al, Rapid Convolutional Algorithm for the Discovery of Blueberry Honey Authenticity Markers via Nontargeted LC-MS Analysis, Analytical Chemistry (2024). DOI: 10.1021/acs.analchem.4c01778
