AI, Meet PD
The prevalence of Parkinson’s is increasing, with estimates that 1.2 million Americans will be living with the disease by 2030, according to a 2018 study by the Parkinson’s Foundation. While most patients know what the disease means for their day-to-day lives, the exact cause of the condition is still a mystery to the medical community. But new breakthroughs in technology are changing the way Parkinson’s is diagnosed and treated—shedding light on the effectiveness of treatments, improving the ability to accurately monitor an individual’s disease progression, and exploring new medication options.
It’s all thanks to advanced computing technology and the ubiquity of smartphones. Artificial intelligence (AI) algorithms can sort through information that would take humans a long time to analyze and can “learn” how to interpret patient data with incredible accuracy. And smartphones allow patients to take their condition into their own hands—literally—by giving them the power to measure, track, and record their symptoms, disease progression, and treatments.
“We hope to give patients the tools they need to better track and understand their condition,” says George Roussos, PhD, a professor of pervasive computing at the University of London who works on several PD projects.
Parkinson’s symptoms are currently assessed using a number of rating scales, including the Unified Parkinson’s Disease Rating Scale (UPDRS). One of Roussos’ projects, cloudUPDRS, utilizes the accelerometer, gyroscope, magnetometer (compass), and touch screen on Android phones to allow patients to take UPDRS tests and track and measure their symptoms at home. Assessments include holding your smartphone flat on the palm of your hand to measure tremors, conducting a series of tapping motions on the screen, and carrying your phone while you walk to measure gait. This data is recorded, and the results are translated into UPDRS’ numerical scale.
While the app is still in clinical trials, it will eventually allow patients to measure their progression on a regular basis and to track how well their medications are working. Doctors will also have access to this data, enabling them to monitor patients more effectively and to tweak treatments accordingly.
“The fact that we don’t have metrics or markers to assess the effect of a new medication accurately is a major hurdle in figuring out what works and doesn’t work,” Roussos says. “AI will help us be much more precise about disease progression. It can help us sort through data quickly and identify patterns that the human eye doesn’t always see. This, in turn, allows us to monitor a treatment’s efficacy.”
Wearable technology and sensors can also be used to track patient data and measure PD symptoms and progression—including watches and gloves that measure tremors, apps that serve up reminders to take medication, and sensors set up in a patient’s home that collect data in an unobtrusive way.
AI and these other cutting-edge technologies can also help doctors personalize treatments—by looking not just at a person’s medical record, but also at their habits, diet, fitness routine, stress levels, and more. “This is a very exciting opportunity,” Roussos says. “Fifteen years down the road, treatment plans could be individualized for the specific symptoms and the specific biomarkers of a particular patient.”