Technology has changed our lives in so many ways. If there wasn’t an Internet, would we be able to do FaceTime, Skype or Zoom calls with co-workers or loved ones during the pandemic? In addition, advances in imaging technology and biometrics has helped in the diagnosis and treatment of many medical conditions. Technology is also a help when it comes to learning about driving techniques, in particular how they are linked to visual impairments in those over the age of 60 and how they are related to car accidents.
First, the advances in imaging technology. Researchers at Duke University developed a way to better diagnosis and monitor eye diseases like glaucoma. They used a combination of optical coherence tomography and deep neural networks that tracks changes in the retinal ganglion cells in the eye.
The ganglion cells are the primary neurons that process and send visual information to the brain. These cells deteriorate and disappear, which leads to blindness, when someone has a disease, like glaucoma. Optical coherence tomography uses light to examine the eye tissue layers in order to diagnose glaucoma and other eye diseases. While optical coherence tomography allows scientists to view the cell layers in the retina, it only shows the thickness of the cell layer, it doesn’t reveal individual ganglion cells. Not being able to see the cells gets in the way of disease diagnosis and tracking of the progression of the disease, since many ganglion cells need to disappear before doctors can see changes in retinal thickness.
That’s where new technology comes in, namely something called adaptive optics optical coherence tomography (AO-OCT). This imaging is sensitive enough to see individual ganglion cells and it is advanced enough to minimize the effect of optical aberrations that happen when examining the eye.
The higher image resolution makes it easier to diagnose diseases, but it also creates such a large amount of data that image analysis becomes a bottleneck and it slows down the use of this in eye and brain research. Researchers came up with a solution by developing a deep learning-based algorithm that can identify and trace the shapes of the ganglion cells from the AO-OCT scans.
Moving from the micro to the macro, the Alabama VIP Older Driver Study, done at the University of Alabama Birmingham used in vehicle sensors and cameras to assess the driving techniques of older drivers. These devices continually record the habits of the drivers for a six-month period.
Since self-reporting of driving behavior and eyewitness reports of accidents are both biased and incomplete, the use of sensors and cameras eliminate these problems and provide an accurate record of how older drivers navigate the roads.
The best thing about the use of the sensors and cameras in the vehicle is that they can detect near-crashes. A near crash is a circumstance that called for a quick evasive maneuver by the driver to avoid an accident. Near crashes happen more frequently and they have causes similar to actual crashes, so having this data helps researchers learn what contributes to accidents in older drivers.
“Using these naturalistic driving techniques in the VIP study we confirmed that contrast sensitivity impairment, slowed visual processing speed and deficits in motion perception elevated crash and near-crash risk in older drivers,” said Cynthia Owsley, Ph.D., professor and Nathan E. Miles Chair of Ophthalmology in the University of Alabama Birmingham Department of Ophthalmology and Visual Sciences and one of the authors of the study.
These two studies show how technology can complement and add to human knowledge, especially when it comes to vision.