AI-Powered Chip Spots Ten Cancers From a Single Blood Test

The quest for a single test that can accurately detect multiple cancers early on has taken a significant step forward. Researchers have created an automated platform that uses an AI-powered chip to analyse a patient's blood serum.

The system works by targeting exosomes, tiny parcels shed by cells into the bloodstream. A specially functionalised chip selectively captures these exosomes, allowing their molecular contents to be scanned to create a unique 'fingerprint'. An artificial intelligence model then analyses these fingerprints with impressive results.

In tests, the platform distinguished cancer samples from healthy ones with 97.4% accuracy and identified early-stage cancers with 97.08% accuracy. It could even classify ten common cancer types with over 93% accuracy.

Crucially, the analysis revealed a molecule called deoxyadenosine triphosphate, which appears to be consistently elevated across various tumours. This finding points towards a potential 'pan-cancer' biomarker, a universal signal that could revolutionise early cancer diagnosis.