My AcademicTutor
Request Support
← Back to Machine Learning in Biostatistics

Module 5 · Machine Learning in Biostatistics

Applied biostatistical ML case studies.

This final module brings the course together through applied health-data case studies. Students move from clinical risk prediction to survival outcomes, high-dimensional omics, missing data, imbalance, fairness and a final R-based project report.

View Module 5 lessons →Open Lesson 1.1Start learning

5

Lessons

0

Open now

R

Project labs

July 2026

Full release

What this module builds

Applied judgement across real modelling problems.

Real case-study thinking

This module moves beyond isolated methods. Students learn how to organise a complete machine learning analysis around a realistic health-data question.

Clinical prediction workflow

Each case study connects outcome definition, predictor timing, model fitting, validation, interpretation and reporting.

Special medical-data problems

Survival outcomes, high-dimensional predictors, missing data, class imbalance and fairness are treated as practical modelling issues.

Output to report

The final module teaches students how to translate R outputs into a cautious, transparent and clinically meaningful report.

Responsible conclusion

Students learn not to overclaim. The final report must state what the model suggests, what remains uncertain and what validation is still needed.

By the end

Students should complete a responsible ML analysis.

Design an applied clinical risk prediction workflow.
Explain why survival prediction requires time-to-event thinking.
Recognise high-dimensional overfitting risks in omics data.
Discuss missing data and class imbalance in health ML.
Evaluate model performance across clinically relevant groups.
Connect R model outputs to written interpretation.
Write a responsible final prediction-model report.
State validation limits, cautions and next steps clearly.

Module pathway

From applied question to final report.

The module follows the complete project cycle: define the clinical problem, prepare data, fit models, validate performance, interpret outputs and write a transparent report.

Step 1

Define

State the clinical question, target population, outcome, prediction horizon and intended use.

Step 2

Prepare

Check predictors, missingness, outcome balance, variable timing and whether the data represent the target population.

Step 3

Model

Fit a baseline model and, where justified, compare it with more flexible prediction methods.

Step 4

Validate

Evaluate discrimination, calibration, threshold behaviour and performance on data not used for fitting.

Step 5

Interpret

Explain the model output in clinical and statistical language, without turning prediction into causal proof.

Step 6

Report

Write a transparent report covering methods, results, limitations, caution and next steps.

Lesson design

The final case-study lessons will be project-led.

Applied case-study lessons based on realistic health-data problems
Detailed notes connecting modelling choices with clinical interpretation
Browser R coding labs planned for every full lesson
Downloadable R scripts and project-style datasets
Prediction outputs, validation tables and interpretation reports
Caution boxes for censoring, missingness, imbalance and fairness
Final project workflow with report-writing guidance

Current release state

Module 5 is open for preview.

Students can see the Module 5 pathway now. The full lessons are locked while they are redesigned with applied R scripts, case-study datasets, validation outputs, interpretation reports and final project guidance.

Start learning →

Module lessons

Study the lessons in order.

All Module 5 lessons currently route to the waitlist. The full release will move from clinical risk prediction to survival, omics, fairness and the final applied R project.

5.1

Lesson

Locked110–130 minRisk prediction

Clinical risk prediction case study

Apply the full prediction workflow to a patient risk example, from target definition and predictor timing to model fitting, validation and reporting.

Risk predictionClinical questionValidation report
Locked until July 2026

Start learning →

5.2

Lesson

Locked110–130 minTime-to-event ML

Survival prediction and censored outcomes

Understand censoring, prediction horizons, survival probabilities, time-dependent validation and why ordinary classification is not enough for survival outcomes.

CensoringSurvival probabilityPrediction horizon
Locked until July 2026

Start learning →

5.3

Lesson

Locked110–130 minHigh-dimensional data

High-dimensional omics and feature selection

Use omics-style examples to study feature selection, high-dimensional predictors, overfitting, penalisation and validation in biomedical prediction.

OmicsFeature selectionOverfitting
Locked until July 2026

Start learning →

5.4

Lesson

Locked100–120 minModel reliability

Missing data, imbalance and fairness

Study how missingness, rare outcomes, class imbalance and unequal performance across groups can affect model usefulness and trust.

Missing dataImbalanceFairness
Locked until July 2026

Start learning →

5.5

Lesson

Locked120–150 minComplete project

Final applied ML project in R

Bring the full course together in an applied R project: define the question, prepare data, fit models, validate performance, interpret results and write a final report.

Full workflowR projectFinal report
Locked until July 2026

Start learning →

Join the waitlist

Get access updates when Module 5 opens in July 2026.

Module 5 lessons are currently locked while they are redesigned with applied project labs, survival and omics examples, missing data checks, fairness interpretation and final report guidance.

Module 5 overview stays open.
All Module 5 lessons remain opening in July 2026.
Lesson 1.1 remains available as the course preview.
Waitlist visitors can request early access or release updates.

Waitlist form

Request access.

Recommended start

Begin with the open foundation lesson.

Lesson 1.1 introduces the course structure: prediction question, R script, model output, interpretation, report writing and responsible modelling caution.

Open Lesson 1.1 →