There are many symptoms and signs of Parkinson’s disease. Biomarkers in people’s earwax, hair and blood can help spot the disease. So can the way a person talks, moves and their mental health. But early diagnosis is complicated as it requires clinical assessments.
A new test devised by a team of researchers in India uses AI to detect Parkinson’s in how people complete a simple drawing test. Handwriting analysis is a non-invasive way of screening for early signs of neurodegenerative conditions based on small variations.
The study with the test was small, with only 66 participants. People drew with a smart pen that recorded both the images and their hand movements. In the study, 31 people had Parkinson’s and 35 did not. They drew spirals and meander patterns. The AI could determine who had Parkinson’s with 98.95 percent accuracy for the meander patterns and 87.74 percent accuracy for the spirals.
The researchers acknowledged that their test size was small and that larger groups are needed. But, they said, “Overall, the study shows the growing role of AI in early and accessible screening of neurological disorders, supporting the ongoing transition toward preventive and precision healthcare. The proposed framework may ultimately contribute to non-invasive, low-cost and remotely accessible neurological screening and clinical decision support.”
What isn’t clear is if this test could aid earlier diagnosis. Everyone in the Parkinson’s group had already been diagnosed. The study did not diagnose anyone in a practical medical setting without more testing.
Right now, the technology is not able to replace neurological tests or serve as a stand-alone diagnostic test. But it’s interesting to see an AI tool confirm a diagnosis. It suggests, in the future, it could be an integral step to a clear and fast diagnosis for conditions that traditionally have a more drawn-out diagnostic process.
Early diagnosis for Parkinson’s can be crucial to helping slow the disease’s progression and improve a person’s quality of life. AI tests may not diagnose the condition early, but any fast-tracking of a diagnosis is helpful.
