Researchers at the Massachusetts Institute of Technology (MIT) have developed a machine-learning model they say can predict whether people at risk for Alzheimer’s disease will experience clinically significant cognitive decline within the next 2 years.
They presented on the model at the recent Machine Learning for Healthcare conference at the University of Michigan in Ann Arbor.
“Accurate prediction of cognitive decline from 6 to 24 months is critical to designing clinical trials,” said Oggi Rudovic, PhD, a researcher in the MIT Media Lab. “Being able to accurately predict future cognitive changes can reduce the number of visits the participant has to make, which can be expensive and time-consuming. Apart from helping develop a useful drug, the goal is to help reduce the costs of clinical trials to make them more affordable and done on larger scales.”
The model consists of two submodels. The first is a population model, which was trained using a dataset of clinically significant cognitive test scores and other biometric data from people with Alzheimer’s disease and healthy individuals. The model used the data to learn patterns that enable it to predict scores on cognitive tests.
The second is a personalized model, which was trained with data from new participants. The model provides score predictions personalized for each patient based on data collected during recent visits.
A “metalearning” scheme automatically chooses whether the population model or personalized model would provide the best prediction based on the data available for analysis, which researchers explained is often limited for patients with Alzheimer’s disease.
“It’s like a model on top of a model that acts as a selector, trained using metaknowledge to decide which model is better to deploy,” Dr. Rudovic said.
In all, the model provided accurate predictions of cognition test scores 6, 12, 18, and 24 months into the future, researchers reported. They hope to partner with pharmaceutical companies next to integrate the model into clinical trials testing treatments for Alzheimer’s disease.