Researchers from the University of Florida’s College of Health and Human Performance Institute have created a new AI system that efficiently tracks the progression of Parkinson's disease using patient recordings of hand movements.
This innovative system, which analyzes quick tapping movements between the thumb and index finger, can detect subtle changes that are often difficult for humans to spot. This has resulted in more accurate and efficient diagnosis and disease monitoring.
"The beauty of this technology is that a patient can record themselves performing the test, and the software analyzes it and informs the clinician how the patient is moving so they can make decisions," explained Diego Guarin, an assistant professor who developed this technology.
"By studying these videos, we can detect even the smallest alterations in hand movements characteristic of Parkinson's disease that might be difficult for clinicians to visually identify," he added.
Parkinson's, affecting millions worldwide, currently has no cure, with its causes still unclear. This progressive disorder impacts the nervous system and body parts controlled by nerves, causing slowness in movement, tremors, body stiffness, speech changes, and more.
As symptoms emerge slowly and without definitive tests, clinicians rely on a series of exercises to identify and evaluate the disorder's severity. However, the current method of rating the disorder is prone to subjective interpretations, potentially leading to inaccuracies.
Enter AI. An automated system can reveal details previously unnoticed. This new AI tool identified the exact time taken to open and close fingers, suggesting that finger-opening movement is delayed in Parkinson's patients compared to healthy individuals.
"This is new information that is almost impossible to measure without the video and computer, showing us that technology can help to better characterize how Parkinson's disease affects movement and provide new markers to help evaluate therapy effectiveness," Guarin explained.
The system was further refined with help from one of the world's largest AI supercomputers, the HiPerGator at the University of Florida. After training, testing, and refining different models with large amounts of video data, the system can run on a smartphone, making it easily accessible to all.
Michael Okun, medical advisor for the Parkinson’s Foundation, said the development was “transformative.”
“The finger-tapping test is one of the most critical elements used for diagnosis and for measuring disease progression in Parkinson’s disease,” Okun said in the university's announcement, noting that AI was able to do what typically required an expert to interpret.
This isn't the first time artificial intelligence has proven handy in combating Parkinson's. Previously, researchers harnessed AI to detect signs of Parkinson's disease up to seven years before clinical diagnosis by analyzing retinal scans.
Additionally, AI's role in accurately predicting critical diseases like Alzheimer's and cancer has already shown promise.
While researchers continue to search for a cure, early detection and reliable tracking could open new possibilities for treating Parkinson's.
Edited by Ryan Ozawa.