A scientist from the O’Donnell Brain Institute says that they found something that could help doctors detect problems with thinking and memory even before you show any clear signs. A special technology listens to your voice, which can help a doctor know if you are having issues with your brain. This was published in an Alzheimer’s Association book about diagnosis and monitoring changes of illness.
A professor of neurology named Ihab Hajjar explained that they were looking into specific changes in language and sound that can be experienced in the early stages of Alzheimer’s disease, but don’t show up easily to family members or doctors.
Researchers used some very complicated computer programs to study the speech patterns of 206 people. Half of them were having mild difficulty with their thinking and memory, and half of them had no problems. The researchers then compared those results with common signs scientists look for when deciding if someone is having issues with their mental abilities.
People enrolled in a program at Emory University in Atlanta took tests to measure their thinking skills. After that, they were asked to verbally describe some artwork for one or two minutes. The description was then studied with the use of artificial intelligence (AI) which measures how well they talked and described the picture, as well as other things like their grammar and how many ideas they had.
The research team looked at the participants’ voices, samples of fluid from the brain and spine, and MRI scans to see how well special tech could correctly detect if someone had mild cognitive impairment or Alzheimer’s disease. They also checked if the tech could tell how bad those diseases were getting over time.
Before technology like machine learning and NLP, it was really hard to understand how people talk. Most of the changes were too small for us to notice. But, Dr. Hajjar found a new method that worked very well in finding out if someone had mild cognitive issues and if they specifically had Alzheimer’s disease – even when other tests did not pick up on it.
In the study, doctors only had to record a patient’s voice for less than 10 minutes. This was much faster compared to other tests that would require several hours to complete.
Dr. Hajjar said that using machines to analyze someone’s voice could make it easier for doctors to spot people who might be at risk of certain conditions. This would allow patients and their families more time to prepare, and give doctors the ability to suggest different treatments quickly.
A team of scientists recently wrote a paper called “Development of digital voice biomarkers and associations with cognition, cerebrospinal biomarkers, and neural representation.” This paper focuses on Early Alzheimer’s Disease which is a kind of dementia. The scientists are trying to find out if there are any speech patterns or ‘biomarkers’ that can help us diagnose this disease. To do this, they pay attention to things like the way someone talks, their brain signals and the chemicals in their bodies. The results could be helpful in determining whether someone has Early Alzheimer’s Disease or not.
Dr. Hajjar invited some researchers from Emory to help him on a project before joining UTSW in 2022. They have received a grant from the National Institutes of Health to collect voice recordings for further study in Dallas, Texas.
Money to help study this subject was given by National Institutes of Health/National Institute on Aging and the Alzheimer’s Drug Discovery Foundation. Dr. Hajjar is part of a group, called “Pogue Family Distinguished University Chair in Alzheimer’s Disease Clinical Research and Care,” which remembers Maurine and David Weigers McMullan.
