Driven to Discover the State of Your Health
The way you zoom around the Capital Beltway or take on traffic in downtown Washington, D.C., may one day be used to monitor your medical conditions.
George Mason University psychology professor Yi-Ching Lee is conducting research to learn how driving can be used to monitor health conditions as part of a four-year study called “Diagnostic Driving: Real-Time Driver Condition Detection Through Analysis of Driver Behavior.”
The study, which also includes researchers from Drexel University, the Children’s Hospital of Philadelphia, and the University of Central Florida, is funded by a $891,135 grant from the National Science Foundation.
Currently, they are studying how teens and young adults with Attention Deficit Hyperactivity Disorder (ADHD) operate cars when they take their ADHD medication versus when they forget a dose or if the medication dosage needs to be adjusted, she said.
Adherence to medication and the role medication plays in driving behaviors are big issues among those with ADHD.
“We want to know how the driving behaviors are different under well-controlled medication and out-of-control medication,” Lee said.
“Once we understand the differences, we can have a monitoring system in the car to keep an eye on the behaviors to capture any onset of symptoms due to medication adherence issues. This approach is essentially moving from reactive health care to a preventive, proactive, and person-centered health care system.”
Participants are being accepted for the first driving simulator study. For more information contact Lee’s Health Behaviors Lab at DrivingStudyGMU@gmail.com or 703-993-5980.
Machine learning techniques are used to pick up patterns in driving behaviors, especially unsafe maneuvers, and to detect nearby traffic, road configurations, and other data. They are also used to make predictions and will be able to quantify differences in driving behaviors, she said.
More than 300 participants will be a part of the study through the use of driving simulators or by driving their own cars outfitted with special sensors and cameras.
“The sensors will collect vehicle control information from the car itself and GPS locations, and the cameras will record traffic situations in the front,” Lee said. “It is critical to know what the driver sees and does; our machine learning techniques rely on knowing everything the driver sees and does.”
Researchers will collect and analyze data by looking for patterns within each participant and across participants in the same groups (ADHD and non-ADHD).
“If successful, our work will lead to transformative changes in how we monitor many types of patients, not only with ADHD, but those with other medical and post-surgical conditions,” Lee said.
Perhaps in the future, machine learning-equipped computers in cars can monitor behaviors. If deviations are detected, then warnings can be generated and feedback sent to care providers, she said.