SMART PEN DEVELOPED TO DETECT PARKINSON’S DISEASE

Fri 06 June 2025:

Developed by researchers at the University of California, Los Angeles (UCLA), the pen features magnetic ink and a special tip that converts hand movements during writing into electrical signals. These signals are then analyzed by AI to detect markers associated with Parkinson‘s.

More than 10 million people worldwide are thought to be living with Parkinson’s, a neurodegenerative disorder with symptoms including tremors, rigidity, slowness of movement and mobility difficulties. While there is no cure, early diagnosis can help those affected access support and treatments earlier.

Handwriting is considered a complex process reflecting coordination between the brain and hand. Previous studies have shown that Parkinson’s significantly affects handwriting. The newly developed pen can precisely measure these changes.

In tests, the researchers compared the handwriting of three Parkinson’s patients with that of 13 healthy individuals, and the AI-supported pen was able to distinguish between them with over 95% accuracy.

Experts describe the device as a practical tool that could be widely used in the diagnosis of Parkinson’s disease.

__________________________________________________________________________

https://whatsapp.com/channel/0029VaAtNxX8fewmiFmN7N22

__________________________________________________________________________

How Does the Diagnostic Pen Work?

The new pen’s tip is made of small neodymium magnets mixed into Ecoflex, a brand of silicone rubber advertised for production of prosthetics and film props. The body contains a reservoir of ferrofluid “ink,” which is surrounded by a barrel with a built-in coil of conductive yarn.

As a user draws or writes with the stylus, deformations in the tip change the magnetic field, and movement of the ferrofluid makes the pen sensitive to acceleration both across a writing surface or in the air. Minute magnetic fluctuations produce a current in the coil, and changes to that current were analyzed rather than the on-paper results of experimental writing or drawing tasks, as is commonly done in today’s neurological assessments.

Participants were asked to perform several tasks, including drawing loops and writing letters. Normalized data was used to train several types of machine learning algorithms, and the best performing analysis came from a one-dimensional convolutional neural network, which reached over 96 percent accuracy in identifying subjects with Parkinson’s.

Current fluctuations in testing were sometimes less than a microampere, and the study version of the pen connected to a current amplifier with a cable. Eventually, the group would like to transfer data wirelessly from pen to computer or smartphone, says Chen.

SOURCE: INDEPENDENT PRESS AND NEWS AGENCIES

__________________________________________________________________________

FOLLOW INDEPENDENT PRESS:

WhatsApp CHANNEL
https://whatsapp.com/channel/0029VaAtNxX8fewmiFmN7N22