Scientists develop tech that can detect early signs of dementia in older drivers
There have been numerous studies on the influence of dementia on driving behavior which indicates that an individual’s driving habits will alter as neurodegeneration leads to cognitive decline. Using this research, a team of scientists is in the preliminary stages of developing algorithms that analyze naturalistic driving data to one day create a non-invasive device or smartphone application that will be able to detect mild cognitive impairment (MCI) and/or dementia in drivers.
The work is based on the results of a long-term study called LongROAD (The Longitudinal Research on Aging Drivers), which tracked approximately 3,000 elderly drivers for up to four years. While the study was in progress, 33 of the participants were diagnosed with MCI and 31 with dementia.
The data was analyzed using solely driving variables such as the percentage of trips traveled within 15 miles of their home, the length of trips starting and ending at home, and minutes per trip. The series of machine learning models employed in the study were able to predict which drivers developed MCI or dementia with 66 percent accuracy.
If, however, other variables such as age, sex, race/ethnicity, and education level were considered, then the accuracy with which the machines were able to detect MCI or dementia went up to 88 percent.
The results are promising, and researchers are confident that an app or software inside a car will be able to constantly monitor driving patterns to alert drivers of cognitive decline before clinical symptoms appear. Early detection will allow for individuals to seek medical attention sooner, which will likely increase the opportunity to delay the worst symptoms of MCI and dementia for as long as possible.
Scientists develop tech that can detect early signs of dementia in older drivers
There have been numerous studies on the influence of dementia on driving behavior which indicates that an individual’s driving habits will alter as neurodegeneration leads to cognitive decline. Using this research, a team of scientists is in the preliminary stages of developing algorithms that analyze naturalistic driving data to one day create a non-invasive device or smartphone application that will be able to detect mild cognitive impairment (MCI) and/or dementia in drivers.
The work is based on the results of a long-term study called LongROAD (The Longitudinal Research on Aging Drivers), which tracked approximately 3,000 elderly drivers for up to four years. While the study was in progress, 33 of the participants were diagnosed with MCI and 31 with dementia.
The data was analyzed using solely driving variables such as the percentage of trips traveled within 15 miles of their home, the length of trips starting and ending at home, and minutes per trip. The series of machine learning models employed in the study were able to predict which drivers developed MCI or dementia with 66 percent accuracy.
If, however, other variables such as age, sex, race/ethnicity, and education level were considered, then the accuracy with which the machines were able to detect MCI or dementia went up to 88 percent.
The results are promising, and researchers are confident that an app or software inside a car will be able to constantly monitor driving patterns to alert drivers of cognitive decline before clinical symptoms appear. Early detection will allow for individuals to seek medical attention sooner, which will likely increase the opportunity to delay the worst symptoms of MCI and dementia for as long as possible.
