An AI-enhanced finger-prick test for Parkinson’s could predict the disease seven years before the onset of symptoms, a study has found. This adds "to an exciting flurry of recent activity towards finding a simple way to test for and measure Parkinson’s."
Following last year's important announcement by The Michael J. Fox Foundation that researchers have discovered a new tool that can reveal a key pathology of the disease: abnormal alpha-synuclein - known as the “Parkinson’s protein” - in brain and body cells. It was hailed as opening a new chapter for research, with the promise of a future where every person living with Parkinson’s can expect improved care and treatments - and newly diagnosed individuals may never advance to full-blown symptoms.
Now, on the other side of the Atlantic, scientists from University College London have teamed up with colleagues at Germany’s University Medical Center Goettingen to train an AI model to spot blood biomarkers in patients with Parkinson’s.
It went on to correctly forecast the disease in 16 out of 72 patients considered at risk of the condition, in one case seven years before symptoms appeared. Of course, the AI tool will continue to be trained (in order to generate better and better analysis) but the discovery already opens up the possibility of using early treatments to slow the onset of Parkinson’s or even stop it in its tracks.
“At present we are shutting the stable door after the horse has bolted and we need to start experimental treatments before patients develop symptoms,” said UCL senior author, Prof Kevin Mills.
Prof David Dexter, director of research at Parkinson’s UK, said the work was a “major step forward”. Adding: “The findings add to an exciting flurry of recent activity towards finding a simple way to test for and measure Parkinson’s."