AI pen could help detect Parkinson's disease from handwriting

Publicly released:
International
Photo by Owen Michael Grech on Unsplash
Photo by Owen Michael Grech on Unsplash

A pen loaded with magnetic ink could help detect early-onset Parkinson’s disease, according to a small study of 16 people. Using magnetic ink, international researchers were able to convert both on-surface and in-air writing motions to electrical signals, and then used AI to analyse these patterns. The pen could distinguish the handwriting of patients with Parkinson’s disease from those without the disease with more than 95% accuracy, according to the team.

Media release

From: Springer Nature

Biomedical engineering: AI pen can detect Parkinson’s disease from handwriting

A pen loaded with magnetic ink can be used to help detect early onset Parkinson’s disease, suggests a study published in Nature Chemical Engineering. Neural network assisted data analysis in the device can identify differences in the handwriting of people with and without the disease and could potentially allow for earlier diagnoses.

Parkinson’s disease is estimated to affect nearly 10 million people around the globe and is the second most common neurodegenerative disease after Alzheimer’s disease. Parkinson’s disease is also the fastest-growing neurodegenerative disease around the world — and diagnoses are thought to be underestimated in low- and middle-income countries due, in part, to a shortage of medical specialists trained to diagnose the disease in these countries. Since symptoms of the disease include tremors, diagnosis is typically based on observing patient motor skills. However, this method lacks objective standards and is typically dependent on the bias of the clinician.

Jun Chen and colleagues developed an approach to diagnosing Parkinson’s disease from handwriting samples taken with a custom-built pen containing magnetic ink. By converting the movements of the magnetic ink to electrical signals from writing on a surface and in the air, the authors demonstrate that with the assistance of a neural network — a method in artificial intelligence that uses a web of interconnected nodes to learn and distinguish between complex patterns — the pen can successfully distinguish the handwriting of patients with Parkinson’s disease from those without the disease with more than 95% accuracy in a small-scale cohort of 16 individuals.

This diagnostic pen could represent a low-cost, accurate and widely distributable technology with the potential to improve Parkinson’s disease diagnostics across large populations and in resource-limited areas. The authors note that future work should expand the tool to larger patient samples and could explore the potential of the tool to track the progression of Parkinson’s disease stages.

Multimedia

Image 1
Image 1
handwriting AI pen

Attachments

Note: Not all attachments are visible to the general public. Research URLs will go live after the embargo ends.

Research Springer Nature, Web page The URL will go live after the embargo lifts.
Journal/
conference:
Nature Chemical Engineering
Research:Paper
Organisation/s: University of California, USA
Funder: J.C. acknowledges the Vernroy Makoto Watanabe Excellence in Research Award at the UCLA Samueli School of Engineering, the Office of Naval Research Young Investigator Award (award ID N00014-24-1- 2065), National Institutes of Health Grant (award ID R01 CA287326), National Science Foundation Grant (award number 2425858), the American Heart Association Innovative Project Award (award ID 23IPA1054908), the American Heart Association Transformational Project Award (award ID 23TPA1141360) and the American Heart Association’s Second Century Early Faculty Independence Award (award ID 23SCEFIA1157587). S.L. acknowledges the National Institute of Health (NS126918) and the Broad Stem Cell Research Center, the Jonsson Comprehensive Cancer Center and California NanoSystems Institute at UCLA. G.C. acknowledges the Amazon Doctoral Student Fellowship from Amazon AWS and the UCLA Science Hub for Humanity and Artificial Intelligence. G.C. also acknowledges the Predoctoral Fellowship from the American Heart Association and The VIVA Foundation (award ID 24PRE1193744). T.T. and J.C. acknowledge the Caltech/UCLA joint NIH T32 Training Grant (award ID T32EB027629). We also acknowledge the careful editing from the UCLA Writing Center for a one-on-one personalized writing consultation.
Media Contact/s
Contact details are only visible to registered journalists.