Enhancing the validity of CAIDE dementia risk scores with resting heart rate and machine learning: An analysis from the National Alzheimer's Coordinating Center across all races/ethnicities
Bibliographic record
Abstract
INTRODUCTION: The clinical utility of dementia prognostic scores has limited validity across diverse populations. This study aimed to enhance the Cardiovascular Risk Factors, Aging and Dementia (CAIDE) model by incorporating resting heart rate (RHR) using a machine learning method across a diverse population. METHODS: We developed CAIDE and CAIDE-RHR models using a random forest algorithm in the National Alzheimer's Coordinating Center (NACC) dataset. Model performances were assessed using area under the receiver-operating characteristic curve (AUC), Matthew's correlation coefficient (MCC), and the Brier score. RESULTS: Incorporating RHR into the CAIDE model significantly improved predictive accuracy across Black African, Asian, White, and Native Hawaiian populations (mean AUC range: 0.80-0.91). However, this improvement was not observed in the American Indian population, where the AUC decreased from 0.87 to 0.84. DISCUSSION: Our findings highlight significant ethnic differences in dementia risk prediction models. These results underscore the need for validating and tailoring dementia risk scores to ensure applicability across diverse races. HIGHLIGHTS: Incorporating resting heart rate (RHR) into the Cardiovascular Risk Factors, Aging, and Dementia (CAIDE) model significantly improves its predictive accuracy for dementia risk across diverse populations, offering a novel addition to dementia risk models. The application of the machine learning technique enhances dementia risk prediction by capturing complex, non-linear relationships among variables. The improved model enables more precise early identification of individuals at risk of cognitive decline, supporting preventive strategies in dementia care. Resting heart rate, a simple and non-invasive cardiovascular measure, is demonstrated to be a valuable predictor for dementia risk, making it practical for clinical application.
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 | 0.017 | 0.030 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.001 | 0.002 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
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".