Machine Learning for the Analysis of Healthy Lifestyle Data: Scoping Review and Guidelines
Bibliographic record
Abstract
Background: Advances in data science and technology have transformed lifestyle research by enabling the integration of multimodal information and the generation of large-scale datasets. Despite the growing interest in machine learning (ML) within health behavior research, significant methodological gaps remain. Objective: The study aims to systematically review the applications of supervised ML algorithms in the analysis of healthy lifestyle data, with a particular focus on the methodological approaches used. The specific objectives are to explore the types and sources of data used for health outcomes, examine the ML processes used, including explainable artificial intelligence (XAI) methods, and review the software tools used. Additionally, this review aims to provide practical guidelines to enhance the quality and transparency of future ML research in health. Methods: Following the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews) recommendations, the search was conducted across PubMed, PsycINFO, and Web of Science, yielding 65 studies that met the inclusion criteria. Results: Most studies (48/65, 74%) integrated multidomain data from physical activity, diet, sleep, and stress. Data sources were split between self-acquired data (33/65, 51%) and health repositories (32/65, 49%). Single-item measurements were common, particularly for physical activity, diet, and sleep. Although 40 of 65 studies used a multimodel approach, random forest was the most frequently applied algorithm. To improve explainability, 22 of 65 (33.84%) studies incorporated specific XAI methods, with 21 using Shapley Additive Explanation values and 1 using local interpretable model-agnostic explanations. R (R Core Team) and Python (Python Software Foundation) were the most widely used software tools, with variation in the libraries used. Conclusions: This review highlights methodological gaps in the application of supervised ML to healthy lifestyle data. The ML workflow should span from data acquisition to explainability, using iterative steps to improve methodological rigor. Although multidomain data collection enhances the understanding of health issues related to lifestyle, representativeness remains limited due to methodological shortcomings in data acquisition. While random forest was the most commonly used algorithm, a multimodel approach is recommended for a comprehensive comparison. Lifestyle components consistently ranked among the top features in studies integrating XAI. Incorporating XAI methods into the ML pipeline can support personalized interventions, provided data collection is accurate. The R metapackage (tidymodels; Max Kuhn and Hadley Wickham) facilitates process evaluation through unified syntax, improving replicability. Methodological and reporting guidelines and a checklist are provided to enhance transparency and replicability in multidisciplinary ML research.
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 | 0.092 | 0.281 |
| Meta-epidemiology (narrow) | 0.003 | 0.002 |
| Meta-epidemiology (broad) | 0.010 | 0.013 |
| Bibliometrics | 0.036 | 0.023 |
| Science and technology studies | 0.002 | 0.004 |
| Scholarly communication | 0.008 | 0.009 |
| Open science | 0.008 | 0.006 |
| Research integrity | 0.010 | 0.006 |
| Insufficient payload (model declined to judge) | 0.010 | 0.003 |
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".