GabyAccording to researchers at Boston University (USA), scientists are still not entirely sure what exactly causes Alzheimer’s disease. However, they do know its consequences and are increasingly better at identifying its early signs, including those that are hidden. human speech .
The university team has developed a new algorithm. artificial intelligence It analyzes the language patterns of individuals with mild cognitive impairment (MCI). It can predict the six-year progression of cognitive disorder from MCI to Alzheimer’s disease with an accuracy of 78.5 percent.
This work by the researchers is a continuation of their previous study, during which they trained the model to detect cognitive impairments using recordings. voices over 1000 people.
What the research showed
The new algorithm developed by the team was trained on transcribed audio recordings of 166 individuals with MCI, aged between 63 and 97 years, the publication reported. Science Alert Since the researchers already knew who had developed Alzheimer’s disease, the machine learning approach could be used to search for signs in transcribed speech that would link 90 individuals whose cognitive function had deteriorated to the point of Alzheimer’s disease .
After training, the algorithm could be applied in reverse: attempting to predict the risk of Alzheimer’s disease based on language sample transcriptions that it had never processed before. Other important factors, such as age and gender, were added for a final assessment.
“One can consider the assessment as the probability that a person’s condition will remain stable or that they will develop dementia,” noted the lead author of the study, informatics specialist Ioannis Paschalidis.
“Wishing to predict what will happen in the next six years, we realized that we could make such a forecast with relatively good confidence and accuracy. This demonstrates the power of AI,” he added.
Why is this important?
Currently, Alzheimer’s disease is incurable. This raises the question: what is the benefit of early detection of the disease if the end result will be the same? Researchers have responded that some treatment methods can help manage Alzheimer’s disease to some extent. And now this management can begin earlier.
Moreover, early detection provides more opportunities to study the disease and its progression, and then develop effective treatments. At the same time, individuals at risk of developing Alzheimer’s disease can participate in clinical trials in advance.
The team is confident that detecting signs of future Alzheimer’s disease using artificial intelligence will continue to develop. This is the type of test that can be conducted quickly and inexpensively even at home – without any special equipment. It requires no injections or biological samples, just a voice recording. In the future, it is likely that the test could be used in the form of a smartphone app.
“If you can anticipate what will happen, you have more opportunities and time to intervene with medication; at the very least, you can try to maintain stability and prevent the progression to more severe forms of dementia,” noted Mr. Paschalidis.
A new method for detecting signs of Alzheimer’s disease will help better understand how it affects a person in the early stages and why it sometimes develops from mild cognitive impairment and sometimes does not.
The results of the study were published in the journal Alzheimer’s & Dementia.
