by Justin Jackson, Phys.org
A team at the University of California, Los Angeles has developed a low-cost diagnostic pen that converts handwriting into electrical signals for early detection of Parkinson’s disease, achieving 96.22% accuracy in a pilot study.
Parkinson’s disease impairs the motor system, leading to tremors, stiffness, and slowed movements that impair fine motor functions such as handwriting. Clinical diagnosis today largely relies on subjective observations, which are prone to inconsistency and often inaccessible in low-resource settings. Biomarker-based diagnostics, while objective, remain constrained by cost and technical complexity.
In the study, “Neural network-assisted personalized handwriting analysis for Parkinson’s disease diagnostics,” published in Nat
Sixteen individuals participated in the pilot study at UCLA Medical Center, including three PD patients and 13 healthy controls.
The pen’s design integrates a silicone-based magnetoelastic tip embedded with magnetic particles, and a reservoir of ferrofluid ink containing nanomagnets. During writing, mechanical stress from hand pressure deforms the tip and causes ink movement, generating shifts in magnetic flux that induce voltage in an embedded coil.
A patient with PD using the diagnostic pen for a writing task. Credit: Nature Chemical Engineering (2025). DOI: 10.1038/s44286-025-00219-5Participants completed handwriting tasks involving line drawing, spirals, and letter writing on paper and in the air. Signals were captured from both methods and processed using several classification models.
A one-dimensional convolutional neural network model achieved the highest accuracy, identifying Parkinson’s disease with 96.22% accuracy across repeated experiments.
Because it captures motor symptoms directly during writing, the device avoids reliance on visual handwriting outcomes or clinician judgment. With its portable format and minimal setup requirements, the pen may be especially valuable for screening in underserved areas, for identifying undiagnosed individuals, and for use in remote or home settings.
With further validation in a larger cohort size, the pen could be used not only for early detection but also for symptom monitoring and data-driven disease management. Authors described the tool as “a low-cost, widely disseminable and reliable technology” for broad deployment in Parkinson’s care.
More information: Guorui Chen et al, Neural network-assisted personalized handwriting analysis for Parkinson’s disease diagnostics, Nature Chemical Engineering (2025). DOI: 10.1038/s44286-025-00219-5
Diagnosing Parkinson’s disease using a soft magnetoelastic smart pen, Nature Chemical Engineering (2025). DOI: 10.1038/s44286-025-00228-4
© 2025 Science X Network
Explore further
New insight into how the human motor cortex encodes complex handwriting
