A team of researchers in the United States think that they found a new way to detect the early signs of cognitive impairment . Machine learning and algorithms can analyze driving behavior to potentially detect the signs of dementia in driving.

The effects of mild cognitive impairment (MCI) and dementia on driving is a topic that has been researched for years. The Alzheimer’s Association that a person with the early signs of Alzheimer’s Disease ( of dementia) can have difficulty driving safe. Some signs can be forgetting how to get to a familiar destination, confusing the pedals and making more errors than an average driver.

Now, researchers are using this kind of driving behavior to detect cognitive impairment. The goal? Use data from the largest study on older drivers in the U.S. and machine learning to detect the signs of cognitive impairment using an app or a device installed into cars. The study is called LongROAD, reports :

The research utilized data from a novel long-term study called LongROAD (The Longitudinal Research on Aging Drivers), which tracked nearly 3,000 older drivers for up to four years, offering a large longitudinal dataset.

Over the course of the LongROAD study, 33 subjects were diagnosed with MCI and 31 with dementia. A series of machine learning models were trained on the LongROAD data, tasked with detecting MCI and dementia from driving behaviors.

The research was funded in-part by the and the full study can be found published to . The study says it’s among the first to take the driving data gathered on older drivers and use it to detect dementia, Alzheimer’s Disease and more through machine learning.

The researchers have found that using driving behaviors alone, the machine learning models could predict MCI or dementia in drivers with 66 percent accuracy:

Although age was the number one factor for detecting MCI or dementia, a number of driving variables closely followed. These include, “the percentage of trips traveled within 15 miles (24 km) of home … the length of trips starting and ending at home, minutes per trip, and number of hard braking events with deceleration rates ≥ 0.35 g.” Using driving variables alone, the models could still predict those MCI or dementia drivers with 66 percent accuracy.

The accuracy is even higher when combining other factors like age and sex. The models had 80 percent accuracy when given only driving behavior and age. At its best, the models had 86 percent accuracy when age, sex, ethnicity, driving behavior and education were all taken into account.

But don’t expect to see an app for this just yet as the research is still in its early stages. The researchers hope that, over time, this study can be used to develop a system that allows your phone or your car to detect the early signs of dementia and other cognitive impairments based on your driving.