Developing an Equitable Machine Learning–Based Music Intervention for Older Adults At Risk for Alzheimer Disease: Protocol for Algorithm Development and Validation
Bibliographic record
Abstract
BACKGROUND: Given the high prevalence and cost of Alzheimer disease (AD), it is crucial to develop equitable interventions to address lifestyle factors associated with AD incidence (eg, depression). While lifestyle interventions show promise for reducing cognitive decline, culturally sensitive interventions are needed to ensure acceptability and engagement. Given the increased risk for AD and health care barriers among rural-residing older adults, tailoring interventions to align with rural culture and distinct needs is important to improve accessibility and adherence. OBJECTIVE: This protocol aims to develop an intelligent recommendation system capable of identifying the optimal therapeutic music components to elicit engagement and resonate with diverse rural-residing older adults at risk for AD. Aim 1 is to develop culturally inclusive user personas for rural-residing older adults to understand their goals and challenges for music-based digital health intervention. Aim 2 is to develop knowledge embedding-based machine learning (ML) models that use music metadata and survey response data to identify optimal therapeutic music components for enhancing engagement and emotional resonance for depression among rural-residing older adults at risk for AD. Aim 3 is to assess acceptability for personalized therapeutic music sessions and ML-based music recommendations with a separate sample. METHODS: Participants (N=1200) will be aged 55 years or older and residing in the United States. In phase 1, participants (n=1000) will receive 5 randomized songs and complete a survey to understand the sentiment, cultural relevance, and perceived benefit for each song. Brief, researcher-created Likert surveys will be used. In phase 2, survey data will be used to develop ML algorithms in collaboration with the University of Massachusetts Amherst Center for Data Science and Artificial Intelligence. These ML models will be integrated into the digital music intervention and tested with a separate sample of 200 participants. Similar to phase 1, participants will be provided with sets of songs generated by the recommendation system based on the target goal (ie, to reduce depression). The recommendation accuracy of the ML algorithm will be assessed using multiple performance metrics, including root-mean-square error and normalized discounted cumulative gain as well as the mean acceptability score with a goal of 85% user acceptability. RESULTS: Participant recruitment is complete for phases 1 and 2 as of June 2025. Data analysis for the results of aims 1, 2, and 3 are underway and results are expected to be published in the fall of 2025. CONCLUSIONS: This protocol seeks to use ML to improve the equitability and accessibility of a digital lifestyle intervention for AD. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): DERR1-10.2196/73711.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.045 | 0.081 |
| Meta-epidemiology (narrow) | 0.003 | 0.002 |
| Meta-epidemiology (broad) | 0.003 | 0.005 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.003 | 0.002 |
| Scholarly communication | 0.003 | 0.002 |
| Open science | 0.003 | 0.003 |
| Research integrity | 0.003 | 0.005 |
| Insufficient payload (model declined to judge) | 0.069 | 0.011 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".