A study has found that observing how a person presses a keyboard can indicate the presence of Parkinson's disease. Early diagnosis of Parkinson's disease in daily life without visiting a hospital can greatly aid treatment.
Researchers from the University of California, Los Angeles (UCLA) announced on the 5th in the international journal "Science Advances" that they have developed an intelligent keyboard that can diagnose Parkinson's disease early.
Parkinson's disease is a condition where dopamine nerve cells that control involuntary muscle movement decrease, causing tremors in the hands and feet and a heavy gait. It is the second most common neurodegenerative disease after dementia. There are nearly 10 million patients worldwide, but there is still no definitive treatment available. The best approach is to detect it early to slow the disease's progression, but there has not been a suitable method for early diagnosis.
Researchers proposed a method to diagnose Parkinson's disease using a keyboard, noting that patients with Parkinson's often have issues with hand and finger movement. Keyboard usage generates various data, including the "flight time" where the hands stay on the keyboard, the "press time" while the keys are being pressed, and the force applied during input. These data can serve as biomarkers for determining Parkinson's disease.
The research team created an intelligent keyboard system with a pressure sensing feature for diagnosing Parkinson's disease. It converts the user's keyboard pressing actions into electrical signals, allowing for tracking and recording of biometric information. This method can measure keyboard input time and pressure.
They also confirmed the potential for early diagnosis of Parkinson's disease through the keyboard. The researchers gathered 15 Parkinson's patients and 12 healthy individuals, asking them to input about 30 words for 2 to 3 minutes using the intelligent keyboard. They developed software capable of identifying Parkinson's patients based on the analyzed data. The experimental results showed that the intelligent keyboard achieved an accuracy of 96.97% in identifying actual Parkinson's patients.
There are various methods for diagnosing Parkinson's disease. The most common approach involves patients self-evaluating their symptoms or functional issues according to the Parkinson’s disease assessment scale. However, this method has the drawback of including subjective elements and making it difficult to assess the disease's progression. While methods like ultrasound and neuroimaging diagnostic techniques exist, they have limitations due to high expenses and the infrequency of actual imaging diagnoses.
The research team noted that the keyboard method for diagnosing Parkinson's disease has overwhelming advantages in terms of portability and convenience compared to other methods. They stated, "After running the app, users can simply connect the keyboard to automatically conduct typing analysis," adding, "The ability to achieve early diagnosis without being constrained by location or requiring expensive equipment or specialized personnel is also a significant advantage."
References
Science Advances (2025), DOI: www.science.org/doi/10.1126/sciadv.adt6631