An artificial intelligence system, developed by researchers at IBM and Pfizer, detects the unseen onset of Alzheimer’s disease by analyzing voice samples from people who are asked to describe what they see in a sketch. The system can predict the onset of the disease even seven years ago, with an accuracy of 70%.
The system uses language processing algorithms to analyze short excerpts from the speech of people who underwent a common cognitive test, used for years to diagnose dementia, in which one must describe a picture where two children steal cookies from the kitchen cabinet behind their mom’s back. The system detects possible grammatical and syntactic errors in speech, which indicate incipient dementia.
The researchers used voice samples from people who participated in the long-running US Francham study, which has continued to track about 5000 people and their families since 1948. The “smart” system analyzed people’s voices when they were 65 years old and predicted who later (up to 85) would develop Alzheimer’s, which could be confirmed, as the relevant data existed over time.
Article continued below
“The key finding is that seven years before the clinical diagnosis, we can say with 70% accuracy that some people will develop Alzheimer’s. Even if you can’t change the course of the disease, you can prepare and regulate your life better,” said IBM Research vice president of research Ajay Royuru.
The Alzheimer’s Society spoke of “encouraging research”, according to the BBC, pointing out that “early diagnosis of dementia is vital for the quality of life with the disease, as, unfortunately, access for people with dementia to appropriate treatment, support and guidance is often delayed. Existing treatments are the more effective, the sooner people start them, and potential new treatments will also be more effective in the early stages of the disease or before symptoms even appear.”
As Fiona Carragher, research director of the Alzheimer’s Society, said, “the new research is an important step in the use of artificial intelligence to have much more timely and accurate diagnoses, although we need to see these methods further tested on larger and more diverse groups of people. However, we desperately need more research of this kind, given that Alzheimer’s disease causes brain changes up to 15 years before symptoms appear.”
Author: Rachel Kaur - Psychologist, dietitian, life coach
Rachel Kaur is a dedicated dietician and psychologist, with a passion for helping individuals achieve holistic well-being. With a dual expertise in nutrition and mental health, she empowers her clients to make sustainable lifestyle changes, fostering physical and emotional health. Rachel's compassionate approach makes her a trusted wellness partner.
Disclaimer: The articles on this web site are purely provided for reasons of advisory. They concern a healthy human organism and do not in any way replace the advice of a doctor. Before making changes to your diet or daily activity, you should first contact your doctor or personal dietitian.
Log in/Register
An artificial intelligence system, developed by researchers at IBM and Pfizer, detects the unseen onset of Alzheimer’s disease by analyzing voice samples from people who are asked to describe what they see in a sketch. The system can predict the onset of the disease even seven years ago, with an accuracy of 70%.
