If you end up with an upset stomach after eating questionable food, you may consider yourself one of the lucky ones.
Foodborne illnesses are a serious global health threat that can cause far more than temporary digestive discomfort.
According to the World Health Organization (WHO), more than 850 million people become sick each year from contaminated food, leading to over 1.5 million deaths annually.
Food Safety Still Relies on the Sniff Test
Despite technology becoming part of nearly every aspect of modern life, food safety still often relies on a simple sniff test to check whether old milk or leftovers are still safe to eat.
However, the human nose is far from a reliable chemical detector, as anyone reaching for Pepto-Bismol the next morning can attest.
To help protect people from contaminated food, engineers at the University of California, Berkeley, have developed an “electric nose,” with their findings published in Science Advances.
Importantly, the technology could be integrated into everyday appliances to detect hidden pathogens and identify signs of food spoilage.
Carla Bassil, an electrical engineer at UC Berkeley and the study’s lead author, says smart refrigerators equipped with sensors could be an ideal application. Such a system could alert users when foods like broccoli are about to spoil or when chicken is nearing the end of its safe storage period.
16 Sensors Give the Electric Nose Its Sense of Smell
The “electric nose” uses 16 sensors, each designed to respond to slightly different combinations of gases, helping determine whether leftover food is still safe to eat.
Bassil compares the sensors to “digital taste buds,” with each one designed to detect a specific stimulus.
“Each of the 16 sensors is coated with a different sensing film. They detect gases by converting chemical reactions between the film and gas molecules into electrical signals,” she explained during a recent presentation.
Using machine learning, the researchers trained the e-nose to identify 16 different food products, reaching an overall prediction accuracy of nearly 93 percent.
The tested foods included fruits and common nut allergens such as walnuts and peanuts. The system tested raw chicken, milk, and eggs left unrefrigerated for 24–48 hours.
The e-nose has advantages over systems using just 2–10 sensors.
Carbon Nanotubes Enable Room-Temperature Operation
The e-nose operates at room temperature using lightweight, strong carbon nanotube semiconductors.
Producing this artificial sensing system is also relatively straightforward. The researchers use drop casting to coat, rinse, and nitrogen-dry the chip.
“The key to scaling the e-nose is depositing multiple sensing materials in one step,” Bassil explains.
However, there are still several ways the technology could be improved.
Improving Accuracy in Detecting Food Allergens
Researchers are improving the e-nose’s ability to distinguish foods, especially tree nuts from peanuts, both common and potentially fatal allergens.
Because of this, the synthetic nose could eventually be adapted to identify allergens as well. It could help reduce the nearly 3.4 million annual US emergency visits caused by food allergies.
The e-nose can currently detect as little as 0.05 grams of isolated walnut, or approximately one-hundredth of a single nut. However, it has not yet been tested in more complicated real-world environments. Detecting traces of walnut in a cake or one spoiled item in a full refrigerator is far more challenging.
The team is also working on a portable e-nose that can connect to a smartphone app. In the future, diners could use digital sensors to scan sushi before eating.
However, the device’s eventual cost and its potential use in low-resource environments are still uncertain. Many foodborne illnesses result from inadequate refrigeration, contaminated water, and unreliable electricity—basic infrastructure needs that must also be addressed.
However, the potential uses of e-nose technology extend far beyond food safety. It could also support biometric systems by detecting health-related chemical signals, much like trained dogs detect diabetes.
Bassil notes that machine learning has transformed sensor technology by improving pattern recognition and making these systems easier to use.
The e-nose can be trained for specific targets and customized for many applications.
Read the original article on: sciencealert
Read more: https://scitke.com/vilnius-great-synagogue-reveals-hidden-jewish-heritage/
