Boston University researchers have developed an innovative AI model that can predict the likelihood of someone developing Alzheimer’s disease within six years, using just the analysis of the person’s speech. This breakthrough, led by Ioannis Paschalidis of the Hariri Institute for Computing, aims to facilitate early interventions, especially crucial given the advent of new drug treatments for Alzheimer’s. Unlike previous methods that relied on expensive procedures, this AI model utilizes speech-to-text technology and basic demographic data to offer an inexpensive and accessible predictive tool. With an accuracy rate of 78.5% in a study involving 166 participants from the Framingham Heart Study, the model represents a significant advancement in the early detection of Alzheimer’s. The research highlights that older women, those with lower education levels, and individuals carrying the ApoE E4 allele are at a higher risk of progressing to Alzheimer’s. This AI-driven approach could revolutionize the early diagnosis and treatment of Alzheimer’s, impacting the lives of millions as the prevalence of the disease is expected to rise significantly in the coming decades.
Read more at East Bay Times…
