2023-04-17 18:30:00
To date, there is no treatment for Alzheimer’s disease. However, the earlier the neurodegenerative pathology is taken care of, the more its progression can be slowed down. This is why early detection is important. And artificial intelligence could offer a solution. Researchers at the University of Texas Southwestern Medical School have developed a program capable of detecting signs of cognitive decline or Alzheimer’s disease by analyzing the words of patients.
Alzheimer’s: voice analysis helps spot cognitive decline
“Our goal was to identify the subtle language and voice changes that are present in the very early stages of Alzheimer’s disease, but are not easily recognized by family members or a patient’s GP. individual”, explains Dr. Ihab Hajjar, professor of neurology and lead author of the article in the journal Diagnosis, Assessment & Disease Monitoring.
The scientist’s team used advanced machine learning and natural language processing programs to analyze the speech of 206 people. 114 volunteers had criteria for mild cognitive decline while the remaining 92 had none. They had to describe for 1 to 2 minutes the works of art presented to them.
“The recorded descriptions of the image provided us with an approximation of conversational abilities that we could study via artificial intelligence to determine motor control of speech, idea density, grammatical complexity, and other speech characteristics. “indicates the specialist in the statement from texas university.
To verify the effectiveness of the AI analysis, the researchers compared the tool’s results with participants’ cerebrospinal fluid samples and MRI scans. Conclusion: the tool can detect early signs of cognitive decline, and more specifically those of Alzheimer’s disease.
Artificial intelligence enables faster screening
“Before the development of machine learning and natural language processing, the detailed study of speech patterns in patients was extremely laborious and often fruitless, as changes in the early stages are often undetectable to the human ear”, says Dr. Hajjar. Thus for him, this technology could offer a faster and easier method of screening for Alzheimer’s disease. Indeed, during the experiment, the tests carried out on the voice recordings only took about ten minutes, whereas traditional neuropsychological examinations generally last several hours.
“If the results are confirmed by larger studies, using artificial intelligence and machine learning to study voice recordings could provide caregivers with an easy-to-perform screening tool for those at risk.”, says Dr. Hajjar. And if this advance still does not make it possible to cure the neurodegenerative pathology, it still has benefits.“Earlier diagnoses would give patients and families more time to plan for the future and give clinicians greater flexibility to recommend helpful lifestyle changes”concludes the expert.
1681788554
#artificial #intelligence #detect #signs #voice