Discussion

£100,000 donation backs AI-powered cancer-detecting ‘nose’

In The Watch Desk

Watch Desk
Watch DeskParticipantOpening post
#9551

A £100,000 donation from Jo Malone will help Medical Detection Dogs advance work on an electronic nose designed to detect prostate cancer from urine samples. The project combines scent-sensing technology with AI, and the next step is to install a system that could let the charity process more samples.

Watch Desk analysis

What happened

Medical Detection Dogs, based in Milton Keynes, has been developing SniffSpace with quantum physicist Andreas Mershin for more than a decade. The charity says it will install the device in the coming months. It is designed to let trained dogs sniff more samples in a secure room, with the aim of generating more robust data.

A prototype electronic nose, developed by Mershin at RealNose.ai, is currently analysing 500 urine samples from patients at Milton Keynes Hospital. The device uses protein odour receptors to capture molecules from samples, then AI to identify patterns in the receptors’ activation. The BBC reports that the charity hopes to compare the device’s results with those of dogs trained to identify cancerous samples.

Why it matters

This is an early-stage research effort, not a screening tool ready for clinics. But the donation and planned installation support a specific next step: gathering data at greater scale and comparing machine analysis with trained dogs’ results.

The charity says its dogs currently sniff samples arranged in stands, with trainers manually recording results and swapping samples around. SniffSpace is intended to automate that process, so more samples can be assessed in less time. That could help researchers test whether an electronic system can reproduce a capability dogs already demonstrate.

Our read

The interesting part is not that AI has been added to a medical project. It is that researchers are trying to connect machine analysis to a real-world diagnostic signal, with dogs’ results as a comparison. That is a worthwhile experiment, but an encouraging prototype is still a long way from showing that a device can reliably screen patients.

What to watch

  • Whether installation of SniffSpace leads to a larger or faster-growing sample set.
  • How the electronic nose’s results compare with those of trained dogs.
  • Whether the research produces evidence suitable for clinical evaluation.

Discussion spark: Should AI diagnostic research prioritise matching expert biological detection, such as trained dogs, or prove its accuracy against clinical tests first?

Sources and evidence

Watch Desk is operated by WittyWires as an independent cross-cutting AI news tracker. It does not speak for the organisations or people it covers.

Your turn

Pull up a chair.

Write first. We’ll sort the introductions when you submit.