It takes a lot of time and money to diagnose Alzheimer’s disease. After performing lengthy neuropsychological examinations in person, physicians should transcribe, review, and analyze each response in detail. But researchers at Boston University have developed a new tool that could automate the process and ultimately allow it to move online. Its computational model based on machine learning can detect cognitive impairment from audio recordings of neuropsychological tests; no face-to-face appointment is required. His findings were published in Alzheimer’s & Dementia: The Journal of the Alzheimer’s Association.
“This approach brings us one step closer to early intervention,” says Ioannis Paschalidis, co-author of the article and Distinguished Professor of Engineering at BU College of Engineering. He says faster and earlier detection of Alzheimer’s could drive larger clinical trials focusing on individuals in the early stages of the disease and potentially allow clinical interventions to curb cognitive decline: “It can form the basis of a tool online that could reach everyone and could increase .the number of people being examined earlier “.
The research team trained their model using audio recordings of neuropsychological interviews of more than 1,000 people at the Framingham Heart Study, a long-running BU-led project that studies cardiovascular disease and other physiological conditions. Using automated online voice recognition tools; think, “Hey, Google!” -; and a machine learning technique called natural language processing that helps computers understand text, caused their program to transcribe the interviews and then encode them into numbers. A final model was trained to assess the likelihood and severity of an individual’s cognitive impairment using demographic data, text coding, and actual diagnoses from neurologists and neuropsychologists.
Paschalidis says the model not only was able to accurately distinguish between healthy individuals with dementia, but also detected differences between those with mild cognitive impairment and dementia. And, it turned out, the quality of the recordings and the way people spoke, whether their speech was constantly advancing or faltering, were less important than the content of what they were saying.
“We were surprised that speech flow or other audio features aren’t so critical; you can automatically transcribe interviews reasonably well and rely on text analysis using AI to assess cognitive impairment,” says Paschalidis, who is also the new director of Rafik B of BU. Hariri Institute of Computer Science and Computer Science and Engineering. Although the team still needs to validate its results with other data sources, the findings suggest that its tool could help doctors diagnose cognitive impairment through audio recordings, including those of virtual appointments or telehealth.
Detection before the onset of symptoms
The model also provides information on which parts of the neuropsychological examination might be more important than others in determining whether an individual has impaired cognition. The researchers’ model divides the transcripts of the examinations into different sections based on the clinical trials performed. They found, for example, that the Boston Naming Test, during which doctors ask individuals to label an image with a word, is the most informative for an accurate diagnosis of dementia. “This could allow doctors to allocate resources in a way that allows them to do more detection, even before the onset of symptoms,” Paschalidis says.
Early diagnosis of dementia is not only important for patients and their caregivers to create an effective treatment and support plan, but it is also crucial for researchers working on therapies to slow and prevent the progression of dementia. Alzheimer’s. “Our models can help physicians assess patients in terms of their chances of cognitive impairment,” says Paschalidis, “and then better tailor their resources by doing additional testing on those with a higher likelihood of dementia.”
Want to join the research effort?
The research team is looking for volunteers to do an online survey and submit an anonymous cognitive test; the results will be used to provide personalized cognitive assessments and will also help the team refine their AI model.
Source:
Magazine reference: