Neuromorphic Olfactory Sensor NoseCore Released: E-Nose Identifies Over 5,000 Odor Molecules
Caltech spin-off Scentient launches NoseCore, a neuromorphic olfactory sensor that mimics human olfactory receptor spike coding to achieve real-time identification of 5,000 odor molecule types.
Neuromorphic Olfactory Sensor NoseCore Released: E-Nose Identifies Over 5,000 Odor Molecules
Caltech spin-off Scentient released the neuromorphic olfactory sensor NoseCore on December 26, the first consumer-grade chip capable of smelling and identifying odors like a human nose.
NoseCore's design is inspired by the human olfactory system. The human nasal cavity contains approximately 400 types of olfactory receptors, each with different sensitivity to different odor molecules. The brain identifies specific odors by integrating all receptor signal combinations. NoseCore integrates 1,024 nanoscale chemical sensing units on a 1-square-centimeter chip, each with different sensitivity to various volatile organic compounds.
In standard odor identification tests, NoseCore successfully identified 5,120 odor molecule types covering food spoilage, gas leaks, and disease biomarkers, with 94.7% accuracy and only 8 milliseconds per identification.
Scentient CEO Priya Sharma said the first application is food safety. Japanese convenience chain FamilyMart is testing NoseCore to detect freshness in bento boxes and rice balls, replacing traditional expiration date labels. In healthcare, Scentient is collaborating with Johns Hopkins Hospital on non-invasive disease screening via volatile organic compounds in breath.
Privacy concerns have emerged as NoseCore could theoretically detect human chemical signals including stress hormones and emotional states. Scentient has disabled emotion detection at the hardware level.
Disclaimer
Content is AI-generated. Do not use it as a basis for real decisions. Do not cite it as factual reporting.