AI detects illness before symptoms arise

Parkinson’s disease is the fastest-growing disease worldwide, according to the professional healthcare publication Medscape. Fortunately, United Kingdom researchers have developed a way to detect it before symptoms arise. 

University College London and University Medical Center Goettingen researchers developed an artificial intelligence that can detect proteins related to this illness.

READ: Alzheimer’s linked to viruses says study

It checks blood samples for these biomarkers and compares them with Parkinson’s disease samples. Professor Kevin Mills, the senior author of the technology’s study, said it can spot this condition with 100% accuracy.

How does the AI detect Parkinson’s disease?

Professor Mills spoke with Interesting Engineering regarding this innovation. He said they started with a study that focused on patients with Rapid Eye Movement Sleep Behavior Disorder (iRBD). 

Roughly 75% to 80% of iRBD patients develop synucleinopathy, a group of brain disorders that includes Parkinson’s disease. 

Then, they developed an AI tool that analyzed those blood samples. It identified that 79% of the iRBD patients’ blood profiles matched those of Parkinson’s patients. 

The researchers followed these patients for ten years while testing the AI tools on their blood samples. Consequently, they identified 16 people who developed this neurological condition years before symptoms emerged. 

The research team enhanced the AI tool with each follow-up period, boosting its accuracy to 100%.

Dr. Michael Bartl, co-first author, explained the significance of spotting these biomarkers with artificial intelligence:

“By determining 8 proteins in the blood, we can identify potential Parkinson’s patients years in advance. This means that drug therapies could potentially be given at an earlier stage, which could possibly slow down disease progression or even prevent it from occurring.”

These biomarkers may help future drug development for Parkinson’s disease. Moreover, the researchers want to simplify the blood test so that the AI can detect symptoms earlier than seven years.