Researchers at the University of California, Berkeley developed an electronic nose that uses 16 gas sensors and machine learning to detect spoiled food and common food allergens with greater sensitivity than the human nose. The device was trained to recognize the scent signatures of fresh and spoiled milk, eggs, chicken, fruits, and nuts, and could one day be integrated into smart refrigerators to alert users when food is nearing spoilage. While the electronic nose can detect tiny amounts of allergens, researchers say it still needs more testing in real-world conditions before it could be used in consumer products. In June, the team showcased their device in Science Advances. The nose uses its 16 gas sensors, each programmed to detect different combinations of compounds associated with spoiled food or allergens, and machine learning trained to recognize scents associated with seven foods: strawberry, blueberry, banana, walnut, hazelnut, cashew, and peanut, as well as raw chicken, milk, and eggs, both fresh and after being left out at room temperature for 24 and 48 hours.