Ohio State Researchers Achieve About 73% Accuracy With Scentronic AI E-Nose

Researchers at The Ohio State University have developed an AI-powered electronic nose and wearable scent delivery system that allows users to identify specific odors with about 73% accuracy. The system, called Scentronic, combines an AI e-nose to detect and analyze odors with a wearable device that releases scents based on machine learning algorithms, marking a significant step toward digitizing the sense of smell.

How the AI E-Nose and Scentronic Work

The new technology uses an artificial intelligence e-nose to detect and analyze odors, which are then paired with a wearable release system called Scentronic. The wearable uses machine learning to classify data in real time, enabling hands-free olfactory stimulation. After being trained on five distinct food scents—coffee, beer, red wine, orange, and chocolate cake—field testing showed participants could identify individual odors about 73% of the time and paired scents about 71% of the time.

This work combines innovations in hardware and software to bring new solutions to societal challenges, said Jinghua Li, senior author of the study and an assistant professor of materials science and engineering at The Ohio State University. Our long-term goal here is to incorporate these systems into compact and personalized devices.

Ohio State Researchers Achieve About 73% Accuracy With Scentronic AI E-Nose — Photo: Ohio State News

Creating Multiple Wearable Versions

The Scentronic system is designed with flexibility in mind, with researchers creating multiple versions for potential consumers. These include a wristband device that users can lift to their nose and a necklace version where five odor channels align like pearls on a string, opening only after a signal is received. The study was recently published in Science Advances.

The team found that Scentronic could be used alongside interfaces that generate taste, demonstrating the feasibility of programmable, cost-effective odor delivery systems for multisensory experiences. Researchers also noted significant individual variation in sensitivity to different scents. We did see that some people are more sensitive to certain odorants than others, Li said. Despite being easy to distinguish, there was still lots of variation in intensity among individuals.

Ohio State Develops Tech To Transmit Smells Over Distance | WION

Supporting Medical Applications

The technology has broad potential applications, particularly in medical contexts. The loss of olfaction is related to many neurological diseases and infections, Li said. As a result, it’s important that we create technologies to first decouple the secrets within our olfactory systems and then try to reconstruct them based on that information.

The system could also enhance extended reality experiences beyond sight and sound, support assistive technologies, and improve virtual training environments. Its versatility suggests it could reduce the need for frequent clinic visits, improve patient comfort, and enable timely, personalized care, according to the researchers.

Digitized Smells May Arrive Soon

The initial demonstrations are promising, so digitized smells may arrive sooner than you think.

The research was supported by the National Science Foundation, the Center for Brain Injury Recovery & Discovery Pilot Award at Ohio State, and the Ohio State Alumni Grant for Graduate Research and Scholarship. Other Ohio State co-authors include Chunyu Yang, Qi Wang, Antonia Akoto-Yeboah, Shaoru Cheng, Billings Amoah, Yizhen Jia, Shulin Chen, Akhil Canumalla, Kai Zhao, and Christopher Simons, as well as Chen Wang from Zhejiang Laboratory in China.