MHDSR
Methods in Health Data Science with R (MHDSR) is a live, cohort-based training pathway in applied biostatistics and machine learning for health research, taught in R with reproducible Quarto workflows.
Enrollment is open for the September–November 2026 cohort. To register, email elbadisyimad@gmail.com with your name and chosen course(s). See the full catalogue →
Programme design
The MHDSR pathway has four courses. BIOSTATR is the base course: R and data manipulation, descriptive statistics, inference, linear and logistic regression, survival analysis, and missing data and multiple imputation, all with reproducible R/Quarto workflows. ML4HOR is the advanced course: supervised learning from classic models to boosting and neural networks, unsupervised learning and clustering, tuning and validation, and interpretability. AIDSR extends the pathway into agentic and AI-assisted data science: driving a reproducible pipeline with a command-line coding agent, and using LLMs for structured extraction and text classification, with guardrails. METAR is a focused course in conducting meta-analysis: review protocol and search strategy, effect sizes, fixed- and random-effects models, heterogeneity, forest plots, meta-regression, publication bias, and certainty of evidence.
Each course can be taken independently, but ML4HOR, AIDSR, and METAR assume the level of R covered in BIOSTATR, and AIDSR additionally builds on ML4HOR.
BIOSTATR → ML4HOR → AIDSR / METAR. Browse the full catalogue and schedule →