Alzheimer’s research: Scientists create AI model to ‘catch Alzheimer’s disease early’

Boston researchers say they’ve created a promising AI model that predicts the likelihood of someone developing Alzheimer’s early.

The Boston University researchers on Tuesday announced that they designed the new artificial intelligence computer program — which identifies those with mild cognitive impairment who are likely to develop Alzheimer’s within six years.

The AI model predicts whether a patient will develop Alzheimer’s just by analyzing the person’s speech.

“Our motivation for this study was to catch Alzheimer’s disease early — opening the door for interventions, which is particularly important now that there are new drug treatments available,” said Ioannis Paschalidis, director of the Hariri Institute for Computing at Boston University.

“If you think about many other chronic diseases including diabetes, hypertension, even cancer, there are early diagnostic procedures and even predictive models that can do prognosis,” he added. “This has not been the case for Alzheimer’s disease, and that’s why we think our study makes an important step in that direction.”

Mild cognitive impairment is a high-risk factor for Alzheimer’s. The condition can be caused by many health factors, but about 3% to 15% of individuals with the condition will progress to Alzheimer’s each year.

While past studies tried to predict the conversion from mild cognitive impairment to Alzheimer’s using expensive methods, the BU scientists created the AI-driven inexpensive method with speech-to-text modeling and basic demographic data.

According to the researchers, this AI predictive method achieved an accuracy of 78.5% in a study with 166 Framingham Heart Study participants.

“Speech is a ubiquitous modality, and it reveals quite a bit about a person’s cognitive status, from sentence completion to the way they structure their thoughts and dialogue,” Paschalidis said.

The researchers used several AI models to process voice recordings of patients captured during neuropsychological exams from the Framingham Heart Study.

The study results suggest that older women with lower education levels and those carrying one or two copies of the apolipoprotein E (ApoE) E4 allele are more likely to progress to Alzheimer’s disease.

This is consistent with earlier research suggesting that people who inherit one copy of ApoE E4 genotype have a high risk of developing the disease, while those who inherit two copies have an even higher risk.

Women who progressed to Alzheimer’s averaged 1.4 years older than men, suggesting that women may be more prone to progression due to their longer life span.

The National Institutes of Health reports that about 6.7 million Americans aged 65 and older are living with Alzheimer’s today, and that this number could grow to 13.8 million by 2060 barring the development of medical breakthroughs to prevent, slow, or cure the disease.

Paschalidis said, “With continued development and refinement, our predictive model may contribute to early intervention and selection in clinical trials for novel AD treatments, ultimately improving patient outcomes.”