Scientists can use driving data and machine learning to detect dementia

The algorithm being used currently has a 66% success rate.

Kyle Hyatt Former news and features editor
Kyle Hyatt (he/him/his) hails originally from the Pacific Northwest, but has long called Los Angeles home. He's had a lifelong obsession with cars and motorcycles (both old and new).
Kyle Hyatt

A new study uses machine learning and senior driving data to detect dementia.

Hill Street Studios/Getty Images

Diagnosing a condition like dementia can be difficult under the best circumstances. But, sometimes, if it goes undiagnosed or ignored, something terrible can happen, particularly if the afflicted person is behind the wheel of a car.

Thankfully, scientists are working on ways to help diagnose dementia earlier, and one of the ways that are being investigated is using machine learning combined with driving data, according to a recent report published by New Atlas.

The study takes vehicle data, including trip length, time of trip from point A back to point A, and how many times a driver brakes hard during the trip and plugs it into an algorithm that helps detect signs of dementia or mild cognitive impairment. The folks behind the study claim that the algorithm has a 66% accuracy rate.

The data for the study comes from a long-term study called LongROAD (Longitudinal Research on Older Drivers), which compiled data from 3,000 senior drivers for up to four years. During the course of the LongROAD study, 33 of the drivers being studied were diagnosed with MCI, and 31 diagnosed with dementia.

So, what does this mean from a practical standpoint? It means that having the conversation with Mom or Dad about taking away their keys is still going to be hard, but you may end up having some data to back you up.

Acura's Type S brand has a rich performance history

See all photos
Watch this: 2022 Volkswagen Taos: Small but mighty