Artificial intelligence could help detect Parkinson's disease, BU study finds

BOSTON - A new study found that artificial intelligence could help scientists identify who is at risk for Parkinson's disease.

Right now, Parkinson's is diagnosed when a patient develops symptoms such as a resting hand tremor or changes in how they walk but there is no simple blood test to say who has Parkinson's and who doesn't, nor who is likely to get it.

A team of scientists in Australia and at Boston University have developed a tool that may help detect Parkinson's disease years before the first symptoms appear. They studied blood samples taken from healthy adults who then developed Parkinson's up to 15 years later and using machine learning, they were able to identify unique chemical markers that could potentially predict whether someone will go on to develop Parkinson's.

Not only that, but the test may also offer clues on how to prevent the disease in these people, such as whether dietary changes could make a difference. 

In:
• Health
• Artificial Intelligence
• Parkinson's Disease

Mallika Marshall, MD

Mallika Marshall, MD is an Emmy-award-winning journalist and physician who has served as the HealthWatch Reporter for CBS Boston/WBZ-TV for over 20 years. A practicing physician Board Certified in both Internal Medicine and Pediatrics, Dr. Marshall serves on staff at Harvard Medical School and practices at Massachusetts General Hospital at the MGH Chelsea Urgent Care and the MGH Revere Health Center, where she is currently working on the frontlines caring for patients with COVID-19. She is also a host and contributing editor for Harvard Health Publications (HHP), the publishing division of Harvard Medical School.

Twitter Facebook