‘E-nose’ Can Detect Fine Whiskey, Counterfeit Perfume

Enter the electronic nose, or “e-nose,” developed by the researchers at the University of Technology Sydney. Associate professor of engineering Dr. Steven Su worked with a small group of PhD students and chemists to create a prototype (internally called NOS.E) that imitates the human olfactory system. It contains eight gas sensors which pick up the odor molecules from a vial of liquid. Depending on the molecules detected, the e-nose’s sensor array develops a unique signal matrix and sends the resulting data to a computer, where a machine learning algorithm calculates the liquid’s characteristics. 

Dr. Su and his team trained their algorithm to identify different whiskey brands, geographical origins, and styles. They had the e-nose “sniff” vials of three blended malts and three single malt whiskeys. The team’s study, published this month in the journal IEEE Sensors, says the e-nose was able to reach accuracy levels of 96.15 percent for the whiskey’s brand name, 100 percent for origin, and 92.31 percent for style classification. 

Unfortunately the e-nose does not look like a human nose. (Photo: UTS)

“An expert can identify the differences between whiskies, but it is difficult for the majority of consumers to differentiate fraudulent beverages,” the study reads. “Complex chemical and analytical analyses have been able to detect the differences between whiskies; however, this type of analysis is time-consuming, complex, requires trained professionals, and can only be conducted in the laboratory.” Conversely, the researchers’ e-nose is said to be capable of detecting six whiskeys’ characteristics in under four minutes—a potential game-changer when it comes to quality control and fraud prevention.

So far the e-nose has been used to differentiate between various types of whiskey, cognac, wine, perfume, and even illegal animal parts, which are frequently sold on the black market. Dr. Su’s team hopes it will eventually find its way to the medical field, where it may assist with disease detection.

Now Read: