AI-driven discovery paves the way for early Parkinson's detection

Researchers, leveraging artificial intelligence, have identified a biological signature for Parkinson's disease, potentially enabling a blood test to detect the condition at least 7 years before symptoms emerge. This groundbreaking predictive test could revolutionise Parkinson's treatment and research, Kazinform News Agency reports.

The progressive disease affects over 150,000 people in the UK, with common symptoms including slowness of movement, tremors, and muscle stiffness. Currently, no drugs can slow or stop Parkinson's, and diagnosis often comes too late to prevent significant brain cell damage.

Researchers from University College London and University Medical Centre in Goettingen utilised machine learning to analyse blood samples from individuals with Parkinson's. They identified 8 key proteins, or biomarkers, consistently present in those with the condition. Their AI tool successfully predicted which patients, originally sampled a decade ago with Rapid eye movement sleep behaviour disorder (a condition with a high probability of progressing to Parkinson's), would develop Parkinson's up to 7 years before symptom onset.

This development marks a crucial step towards improving early diagnosis and treatment of Parkinson's, offering hope for better management of the world's fastest-growing neurodegenerative disorder.