Quantum sensors can detect tiny biological signals, but they currently need expensive lab conditions and hours of manual analysis—AI could make them fast, cheap, and portable enough for a GP surgery or a rural clinic. This project tackles a specific bottleneck: even the most sensitive quantum sensor fails in messy real-world environments, like a busy hospital ward or a patient’s bedside. The researchers will train AI to cut measurement times, filter out noise, and automate image analysis. In one case, they have already shown that spin-enhanced nanodiamond sensors can detect biomarkers 100,000 times more sensitively than current lateral flow tests—but only under ideal lab conditions. AI is the tool to make that sensitivity work in a real blood sample. If successful, the work could transform diagnostics for epilepsy, cancer, and infectious disease. Portable quantum sensors could replace bulky, expensive hospital equipment, bringing ultra-sensitive testing to low-resource settings. The project also includes training bootcamps to spread AI expertise across the UK quantum research community, accelerating adoption beyond this single grant.
View original technical description
We are developing a new generation of quantum sensors to enable earlier diagnosis through ultra-sensitive tests to detect lower biomarker levels, and improved access through lower costs and portable instruments. Despite major breakthroughs, quantum sensors still face barriers before their full potential for healthcare applications can be realised by deploying them in in hospitals, clinics, or low-resource settings. Current experiments can require long measurement times, carefully controlled lab environments, and expert manual analysis. These constraints threaten to prevent widespread use in real diagnostics. Even where the fundamental sensitivity is high, turning that sensitivity into real clinical performance (particularly in messy, variable environments) is an ongoing challenge. There is a pressing need to integrate artificial intelligence (AI) to make quantum sensing technologies faster, smarter, and more adaptable to complex real-world conditions. This project brings together researchers from various disciplines and UK institutions to use AI to address these challenges for quantum sensors in five key areas: Making quantum sensors smarter and faster: We will use AI to reduce the number of measurements needed, by identifying and focusing only on the most informative data points. This will reduce scan time and improve performance in noisy and unpredictable conditions. Boosting in vitro diagnostics: We are developing spin-enhanced lateral flow tests that use nitrogen-vacancy centres in nanodiamonds, and have demonstrated a 100,000-fold improvement in fundamental sensitivity over the current standard method. To translate this into real-world assays, we will use AI for automated image analysis that can improve sensitivity and distinguish between specific and nonspecific binding. Improving epilepsy detection: We will use AI to automatically detect epileptic brain activity from traditional magnetoencephalography and quantum (optically pumped) magnetoencephalography. Discovering better materials for diagnostics: Using AI to explore and optimise new fluorescent nanomaterials for use in blood-based cancer monitoring. This will allow faster, more accurate detection of circulating tumour DNA in blood samples. Driving the wider adoption of AI: Hosting training courses and bootcamps for researchers interested in learning how to apply AI techniques for their own research.
Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.
Is something wrong? Let us know