OdorPi Secures £700K to Build AI-Powered ‘Electronic Nose’ for Diagnostics
London-based deeptech startup OdorPi (OdorPi is now Anemo Labs) has secured £700,000 in pre-seed funding to commercialize technology that gives artificial intelligence a sense of smell. The round was co-led The post OdorPi Secures £700K to Build AI-Powered ‘Electronic Nose’ for Diagnostics appeared first on Ventureburn .
London-based deeptech startup OdorPi, now operating under the name Anemo Labs, has secured £700,000 in pre-seed funding to bring its AI-powered electronic nose to market. The investment round was co-led by Zinc and SFC Capital, with additional backing from Innovate UK. Founded in 2025 by Koye Sodipo, OdorPi is creating an electronic nose that fuses sensor hardware with machine learning algorithms.
The system aims to identify and classify volatile organic compounds (VOCs), the chemical substances that generate scent. By analyzing these compounds in urine, OdorPi hopes to detect disease markers before physical symptoms appear. The company has engineered a proprietary sampling platform capable of gathering and labeling up to 5,000 smells per day to train its models.
Initial testing using the Sniffin’ Sticks olfactory protocol yielded an 84.15% accuracy rate when classifying 12 different scent categories. Planning to utilize the new funds, Anemo Labs will focus on enhancing sensor chemistry and augmenting its collection of smell data. CEO Sodipo highlighted olfaction as a non-invasive, continuous approach to disease screening, with potential future uses for urinary tract infections and certain types of cancers.
The scientific team, comprising experts in electronics and sensor chemistry, is currently collaborating with clinical advisors to validate the diagnostic platform. While the £700,000 amount is relatively small compared to other deeptech funding rounds, it demonstrates investor confidence in transforming a traditionally manual clinical observation method into a digitized, scalable diagnostic tool.
Written by urgent.news from Ventureburn's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.