Scientists at Boston University have developed an innovative AI model capable of predicting Alzheimer’s disease progression with 78.5% accuracy by analyzing patients’ speech patterns. This breakthrough could lead to earlier diagnoses and more accessible screenings, significantly impacting dementia care worldwide.
Alzheimer’s disease diagnosis could soon become more accessible and accurate thanks to a groundbreaking artificial intelligence (AI) model developed by researchers at Boston University. The newly designed AI computer program predicts, with high reliability, whether individuals with mild cognitive impairment are likely to develop Alzheimer’s-related dementia within the next six years. This innovative approach, which leverages speech analysis, has the potential to transform early diagnosis and intervention efforts for Alzheimer’s disease.
This AI model predicts, with an accuracy rate of 78.5%, whether someone with mild cognitive impairment will remain stable or transition to Alzheimer’s dementia over the next six years,
Traditional diagnostic methods for Alzheimer’s often involve extensive assessments, including interviews, brain imaging and various lab tests. These methods can be time-consuming and are typically only performed once noticeable symptoms manifest. By then, the disease may have already caused irreversible damage.
The research, published in Alzheimer’s & Dementia, the journal of the Alzheimer’s Association, details the development and testing of this AI model. The project taps into data from the renowned Framingham Heart Study, which, though primarily focused on cardiovascular health, also collects extensive cognitive functioning data from participants showing signs of cognitive decline.
Researchers used audio recordings of 166 clinical interviews with individuals aged 63 to 97 diagnosed with mild cognitive impairment. They applied speech recognition technology similar to that used in smart speakers, along with machine learning algorithms, to identify patterns linking speech characteristics to cognitive outcomes.
“We found we can reasonably make that prediction with relatively good confidence and accuracy,” Ioannis (Yannis) Paschalidis, director of the BU Rafik B. Hariri Institute for Computing and Computational Science & Engineering and a founding member of BU’s Faculty of Computing & Data Sciences, said in a news release.
This capability enables clinicians to anticipate the progression of Alzheimer’s and introduce early interventions, potentially stabilizing the patient’s condition.
The practical implications of the research are profound. This model could make cognitive impairment screening and early diagnosis more accessible by automating significant portions of the process, thus reducing the reliance on expensive and specialized tests.