We still don’t know exactly what causes Alzheimer’s, but we do know what its effects are, and we’re getting better at detecting its early signs, possibly even during speech.
Scientists at Boston University have developed a new AI (artificial intelligence). algorithm The system, which analyzes the speech patterns of patients with mild cognitive impairment (MCI), can predict with 78.5% accuracy whether MCI will progress to Alzheimer’s disease within six years.
This research is the team’s Previous researchSo the researchers used voice recordings from over 1,000 people to train a model to accurately detect cognitive impairment.
Their new algorithm was trained on transcribed voice recordings of 166 people with MCI, aged between 63 and 97. The team already knew who had developed Alzheimer’s, so Machine Learning This approach is Their transcribed speeches It linked 90 people to cognitive decline and Alzheimer’s disease.
Once trained, the algorithm could then be used in reverse to predict Alzheimer’s risk from transcripts of previously unprocessed voice samples, with other important factors like age and self-reported gender added in to arrive at a final predictive score.
“The score can be thought of as the likelihood, the probability, that someone will remain stable or progress to dementia.” To tell Ioannis Paskalidis, a computer scientist at Boston University.
“We wanted to predict what’s going to happen over the next six years, and we found that we could reasonably make that prediction with a relatively high degree of confidence and accuracy. This shows the power of AI.”
Given that there is currently no cure for Alzheimer’s disease, you might wonder what the benefit of early detection is when the outcome is the same. However, there are treatments. Helps manage Alzheimer’s To some extent, these can start sooner.
In addition, early detection provides more opportunities to study the disease and its progression, from which a fully effective treatment can be developed. People who are known to be at high risk of developing Alzheimer’s can participate in clinical trials in advance.
This approach would be highly valued if it were further developed: it is a test that can be performed quickly and cheaply at home without specialist equipment, no injections or samples required, just a record, and in the future, it may be possible to do it via a smartphone app.
“If we can predict what’s going to happen, we have more opportunity and time to intervene with drugs, at least to keep the disease stable and prevent it from progressing to more severe dementia.” To tell Pashalidis.
The recordings used here were quite rough and of low quality, and with cleaner recordings and data the algorithm would likely be even more accurate.
This may lead to a better understanding of how Alzheimer’s affects us in the very early stages, and why some people develop Alzheimer’s disease from MCI and others don’t.
“Like everyone else, we hope for more and more treatments for Alzheimer’s.” To tell Pashalidis.
This study Alzheimer’s and Dementia.