New Software May Help Detect Early Parkinson’s Motor Signs At Home

Researchers have validated a software that evaluates typing patterns with keyboards to detect Parkinson’s disease-specific motor impairment. This approach, done in an at-home setting, may allow early detection of the disease, as well as monitor disease progression.

The study, “Detecting Motor Impairment in Early Parkinson’s Disease via Natural Typing Interaction With Keyboards: Validation of the neuroQWERTY Approach in an Uncontrolled At-Home Setting,” was published in the Journal of Medical Internet Research.

Early detection of Parkinson’s disease can be crucial to prevent disease progression. The current standard to evaluate motor signs is the Unified Parkinson’s Disease Rating Scale part III (UPDRS-III), which requires a trained specialist and attendance at the clinic. This limits the frequency at which disease state and progression can be assessed.

Recently, efforts have been made to develop more accessible methods to detect motor signs of Parkinson’s, including the use of digital technologies. Analysis of the time a person takes between pressing and releasing a key while typing (key hold time) was found to be a reliable method to detect impaired psycho-motor function.

A recent study showed that analysis of key hold times could detect motor signs in the early stages of Parkinson’s in a controlled typing task in the clinic. The work involved a new computational algorithm able to generate a Parkinson’s disease motor index based on key hold times, called “neuroQWERTY index” (nQi).

This approach measures the key hold times during the normal use of a computer without any change in hardware and converts it to a neuroQWERTY index. This has the potential to detect motor problems remotely, in a natural environment (like home),  which would allow data to be collected much more often than current standard of care.

Researchers evaluated the use of the neuroQWERTY approach in an uncontrolled at-home setting. This study analyzed the baseline data collected from participants who had less than five years of disease and were about to initiate dopaminergic therapy, in a six-month Parkinson’s clinical trial (NCT02522065).

At the beginning of the trial, 60 participants (30 early-diagnosed Parkison’s patients and 30 healthy controls) underwent clinical evaluation, that included the UPDRS-III method and the neuroQWERTY typing test in clinic (which takes approximately 15 minutes).

Participants who reported using the computer for at least 30 minutes a day had the platform and software installed in their personal laptops. Data was collected for seven days after the first log-in to the neuroQWERTY platform, and participants were encouraged to type an email or a document for at least 15 minutes per day.

At the end, only 52 participants had enough data for the final analysis — 25 Parkinson’s patients and 27 healthy individuals. Researchers compared that data with the one collected during the typing task in the clinic.

The neuroQWERTY approach at home was able to distinguish Parkinson’s patients from healthy individuals through the analysis of at-home typing patterns, and had a comparable performance to that performed in the clinic.

“These results prove that the data collected from subjects’ routine use of the computer also are valid to detect PD-related motor signs, getting us closer to our ultimate goal of providing an objective ambulatory tool to monitor PD progression,” researchers wrote.

The team now wants to develop a tool that can track Parkinson’s progression and therapeutic effectiveness. However, additional studies must be performed to validate the neuroQWERTY approach to monitor Parkinson’s disease progression over time.