A simple blood test using artificial intelligence to predict Parkinson's disease years before symptoms begin has been developed by researchers. They hope it can lead to a cheap, finger-prick test providing early diagnoses - and help find treatments to slow down the disease. Charity Parkinson's UK said it was "a major step forward" in the search for a non-invasive patient-friendly test.
But larger trials are needed to prove its accuracy. Parkinson's affects nearly 10 million people worldwide and more than 150,000 in the UK. Many are diagnosed having developed symptoms, such as tremors, movement and memory problems - caused by nerve cells dying in the part of the brain that controls movement.
There is no cure, and no treatment to slow or stop the disease, although therapies exist to help manage symptoms. The researchers, led by scientists at University College London and University Medical Center Goettingen, in Germany, took blood samples from a group of people with Parkinson's and another without, to identify eight major proteins most likely to predict who would develop the condition. They then tested blood from 72 patients at risk of brain disorders such as Parkinson's for those eight protein markers - and followed them up over 10 years.
Using the AI tool, they correctly predicted 16 would develop Parkinson’s - up to seven years before symptoms began. In total, the test predicted 79% would develop the disease - and the researchers are still following up other patients to co.
