London-based deep tech startup Anemo Labs has raised £700,000 in pre-seed funding to advance an electronic nose that detects disease through smell. The round was led by Zinc and SFC Capital, with additional support from Innovate UK. Founded in 2025 by Koye Sodipo, the company is initially applying its technology to healthcare by analysing volatile organic compounds in urine for non-invasive screening.
How the Electronic Nose Works
Anemo combines sensor chemistry and electronics with machine learning models trained to distinguish between different smells. Its scientific team includes researchers with PhDs in sensor chemistry and electronics, supported by clinical advisors for early validation. A central element is an in-house sampling platform designed to collect and label up to 5,000 smells per day, providing training data for its classification models.
Early Testing and Accuracy
In early testing based on the Sniffin' Sticks olfactory protocol, Anemo reports that its models correctly classified scent categories 84.15 percent of the time across 12 categories. This result is an early-stage technical demonstration rather than a clinical finding on patient samples. It suggests the system can identify everyday odours with promising reliability while more demanding clinical validation remains ahead.
Funding Allocation and Clinical Targets
The new funding will be used to further develop Anemo's sensor chemistry and expand its smell data collection and labelling capabilities. The company will also continue validating its olfactory diagnostics platform in partnership with clinical collaborators. Longer term plans include diagnostic applications for urinary tract infections and certain cancers where changes in smell can provide a clinical signal.
Founder's Vision
Founder Koye Sodipo described olfaction as a non-invasive, passive and continuous medium to screen for a number of diseases. He said the mission is to enable clinicians and individuals to catch disease even before symptoms appear. This vision underpins the company's focus on urine as a practical first matrix because collection is routine and the sample headspace can act as a repeatable smell source.
Building a Proprietary Dataset
Unlike imaging or text-based AI, there is no large pre-existing public dataset of labelled VOC signatures for medical olfaction. Anemo's response is to build a proprietary dataset through its sampling platform, which is designed to produce labelled smells at scale. The company sees this dataset as central to improving its models and generalising across a wider range of conditions over time.
An Early Stage Technology
Electronic nose technology has remained largely outside routine clinical practice for decades despite promising research. Anemo's current results come from a standard smell identification test rather than patient samples or confirmed disease labels. The company has not named hospital sites or published clinical protocols, and the diagnostic capability remains downstream of further validation.
Potential Clinical Impact
Non-invasive, continuous screening could address a real gap in early disease detection, particularly for conditions that are diagnosed only after symptoms progress. Urine is a practical first sample because it is already collected routinely and provides a repeatable smell source. This approach may reduce reliance on needles and laboratory assays if Anemo can validate its technology across broader patient populations.
Anemo Labs represents an early-stage but novel direction in healthcare AI, using smell as a sensing modality rather than imaging or genomic data. If clinically validated at scale, non-invasive VOC screening could support earlier disease detection for conditions that are often diagnosed late. The pre-seed round provides resources for sensor development, data expansion and partner validation, but the technology still needs to prove itself in real clinical settings.