Technology that analyzes everyday actions such as handwriting or keyboard input to diagnose Parkinson's disease early is being developed one after another. Following the introduction of a system by American scientists that diagnoses the disease by analyzing keyboard input patterns, researchers have also identified the presence of Parkinson's disease through subtle movements in handwriting.
Jun Chen, a professor at the University of California, Los Angeles (UCLA), and his research team noted on the 3rd that they succeeded in early diagnosis of Parkinson's disease using a pen with magnetic ink and artificial intelligence (AI) technology. The research results were published in the international scientific journal Nature Chemical Engineering on the same day.
Parkinson's disease is a degenerative neurological disorder characterized by the reduction of dopamine nerve cells responsible for involuntary muscle movements, resulting in tremors in the hands and feet and a heavy gait. Approximately 10 million people are affected worldwide. The main drawback of diagnosing the disease has been that it relies on visual observation of the patient's movements and tremors, leading to lower accuracy and consistency. In particular, low-income countries often have a low diagnostic rate due to a lack of specialized personnel.
The research team paid attention to the fact that writing requires intricate cooperation between the brain and hand. This cooperation is impaired in Parkinson's disease patients, leading to changes in their handwriting. The researchers first developed a pen containing magnetic ink. When writing, the pen tip is pressed or moves, changing the internal magnetic field, which allows them to collect data on subtle tremors, pressure, and movements of the hand during handwriting.
Subsequently, they conducted clinical trials with 16 participants, including three Parkinson's disease patients. While performing handwriting tasks like drawing waves and spirals and writing uppercase letters using the pen, they analyzed the data with AI, which was able to distinguish the subtle differences in handwriting between Parkinson's disease patients and the general population. The AI identified the presence of Parkinson's disease with an accuracy of over 95% through handwriting.
There are various methods for diagnosing Parkinson's disease. The most common method is for patients to assess their own symptoms or functional problems based on the Parkinson's disease rating scale. However, this approach reflects subjective factors and has difficulties in understanding the progression of the disease. While there are also ultrasound and neuroimaging diagnostic methods, they are often expensive, and there are limitations in the actual cases of imaging diagnosis.
The research team stated, "The magnetic pen is inexpensive and portable, showing potential for effective use in areas with insufficient medical infrastructure," and added, "We plan to expand experiments to more people and explore whether this pen can also track the progression of the disease." The magnetic ink pen can be mass-produced at a low cost using a 3D printer.
Previously, Professor Jun Chen's research team developed an intelligent keyboard that can diagnose Parkinson's disease early using only the action of pressing keys in April. This method analyzes various data such as the time, pressure, and duration of hand movements while pressing keys, utilizing the characteristics of Parkinson's disease, where hand and finger movements become impaired. They reported that, based on experiments with actual Parkinson's disease patients and the general public, they were able to distinguish the disease with an accuracy of about 97%.
This keyboard diagnostic method is also much more convenient and less burdensome in terms of expense compared to existing imaging diagnosis or self-assessment methods. At that time, the research team explained, "It is advantageous that early diagnosis can be achieved without being restricted by location, expensive equipment, or specialized personnel," stating that after running the app (application program), users can connect the keyboard for automatic typing analysis.
References
Nature Chemical Engineering (2025), DOI: https://doi.org/10.1038/s44286-025-00219-5
Science Advances (2025), DOI: https://doi.org/10.1126/sciadv.adt6631