Developing Equitable Machine Learning-Based Music Intervention for Alzheimer's Prevention

Strategic Insights -

Researchers at the University of Massachusetts Amherst are working on a pioneering project to develop a machine learning-based music intervention to prevent Alzheimer's disease among rural-residing older adults. The study aims to create an intelligent recommendation system to identify the optimal therapeutic music components to elicit engagement and resonate with diverse older adults at risk for AD. The research focuses on developing culturally inclusive user personas and machine learning models to identify optimal music components, and assessing acceptability for personalized therapeutic music sessions. The project has already completed participant recruitment for phases 1 and 2, with data analysis underway and results expected to be published in the fall of 2025.

Key Takeaways:

  • The study aims to develop an intelligent recommendation system to identify therapeutic music components for Alzheimer's prevention among rural-residing older adults.
  • The research involves developing culturally inclusive user personas to understand the goals and challenges of rural-residing older adults in music-based digital health interventions.
  • Machine learning models will be used to identify optimal therapeutic music components based on music metadata and survey response data.
  • The study will assess acceptability for personalized therapeutic music sessions and ML-based music recommendations with a separate sample of 200 participants.
  • The project aims to achieve 85% user acceptability for the personalized music intervention.
  • The research protocol is focused on developing an equitable machine learning-based music intervention for older adults at risk for Alzheimer disease.

Statistics:

  • The study will include 1200 participants aged 55 years or older residing in the United States.
  • 1000 participants will receive 5 randomized songs and complete a survey to understand the sentiment, cultural relevance, and perceived benefit for each song.
  • The study will use multiple performance metrics to assess the recommendation accuracy of the ML algorithm, including root-mean-square error and normalized discounted cumulative gain.
  • Participant recruitment is complete for phases 1 and 2, with an expected completion date of fall 2025.

Sources:

  • NewsRx. Researchers from University of Massachusetts Amherst Report Findings in Personalized Medicine (Developing an Equitable Machine Learning-Based Music Intervention for Older Adults At Risk for Alzheimer Disease: Protocol for Algorithm Development and Validation). Drug Week. August 29, 2025; p 4910.
  • Developing an Equitable Machine Learning-Based Music Intervention for Older Adults At Risk for Alzheimer Disease: Protocol for Algorithm Development and Validation. JMIR Research Protocols, 2025;14.