ML4HOR
Machine Learning for Health Outcomes with R
ML4HOR
Machine Learning for Health Outcomes with RPrediction
Course description
ML4HOR provides a practical introduction to machine learning for health outcomes research using R. The course covers the complete supervised learning workflow and explores clustering as a final extension for patient phenotyping, subgroup discovery, and exploratory health-data analysis.
The course emphasizes validation, data leakage prevention, appropriate performance metrics, calibration, interpretability, clinical relevance, and reproducible reporting.
Learning objectives
By the end of ML4HOR, participants will be able to:
- formulate machine learning questions in health outcomes research;
- prepare tabular health datasets for ML analysis;
- fit supervised learners;
- evaluate models using resampling-based validation;
- tune and compare learners fairly;
- assess discrimination, calibration, and clinical usefulness;
- interpret ML models using global and local explanation methods;
- apply clustering for patient phenotyping and subgroup discovery;
- distinguish prediction, clustering, and causal estimation;
- produce a reproducible ML report with Quarto.
Schedule
| No. | Topic |
|---|---|
| 1 | Introduction to ML for health outcomes & R/Quarto workflow |
| 2 | Resampling and performance metrics |
| 3 | Linear and regularized models |
| 4 | Tree-based models, random forests, boosting & ensembles |
| 5 | Hyperparameter tuning, model selection & benchmarking |
| 6 | Calibration and clinical utility |
| 7 | Model interpretation, explainability & clustering for phenotyping |
| 8 | End-to-end ML pipeline and mini-project |
Practical information
- Audience
- Researchers, PhD students, clinicians, public health professionals, data analysts
- Prerequisites
- Basic R and statistical literacy recommended, as covered in BIOSTATR
- Format
- Live sessions, slides, notebooks, datasets, exercises
- Schedule
- Mon/Wed/Fri 18h00–20h00, Sun 10h00–12h00 (Morocco time)
- Tools
- R, RStudio, Quarto
- Assessment
- Reproducible Quarto mini-project
- Next step
- SURV4CR or METAR