MEDS ML4H 2025 Tutorial
Building a model with MEDS: from the key design principles through data extraction, task definition, and model building.
This page hosts the complete form of the notebooks explored interactively at the MEDS ML4H 2025 Tutorial.
What you'll build
Over five notebooks, you'll take the openly available MIMIC-IV demo dataset from raw tables to trained, evaluated models for one running prediction task: will a patient's ICU stay last longer than 3 days? Each step uses a tool from the MEDS ecosystem. A final step shows how to compare your models with community baselines in MEDS-DEV.
| Step | Page | What you'll do | Main tools |
|---|---|---|---|
| 1 | Getting Familiar with MEDS | Learn the MEDS data and metadata files on a small example dataset | The MEDS schema |
| 2 | Converting to MEDS | Convert the raw MIMIC-IV demo tables into MEDS | MEDS-Extract (built on MEDS-Transforms) |
| 3 | Extracting a Prediction Task | Turn the clinical question into a precise task definition and extract its labels | ACES |
| 4 | Build a Tabular Baseline | Build features, train a logistic-regression baseline, and evaluate it | scikit-learn, MEDS-Evaluation (with MEDS-Tab shown as the scalable alternative) |
| 5 | Build a Neural Network | Pre-process the data and train a neural network directly over MEDS data | MEDS TorchData, Lightning |
| 6 | Compare with MEDS-DEV | Relate your models to MEDS-DEV baselines, run them yourself, and contribute | MEDS-DEV |
How to follow along
- Just reading? The pages for steps 1–5 render their full notebooks, outputs included, so you can follow the whole tutorial without running anything.
- Want to run it? Use the "Open in Colab" link at the top of each notebook page, or clone the tutorial repository and run the notebooks locally.
- Start anywhere. Steps 3–5 begin by downloading the MIMIC-IV demo in MEDS format, and steps 4 and 5 also download the task labels produced in step 3, so you can jump straight to the step you care about.
- No credentialed data access needed. Everything uses the open-access MIMIC-IV demo.
Tutorial steps
Getting Familiar with MEDS
Before we start building datasets and models using the MEDS ecosystem, we need to understand what the MEDS
Converting to MEDS
In this tutorial, we'll extract a dataset in MEDS format for downstream use. See below for the jupyter
Extracting a Prediction Task
To easily extract prediction tasks from MEDS Data, we will use the
Build a (Tabular Baseline) Model over MEDS data
In this tutorial, we'll use the dataset we extracted previously and build a tabular baseline model to predict
Build a (Neural Network) Model over MEDS data
In this tutorial, we'll use the dataset we extracted previously and build a neural network model to predict
Compare with MEDS-DEV
Relate the tutorial's models to MEDS-DEV baselines, run MEDS-DEV yourself, and contribute your task, model or dataset.
Tutorial Presenters
Additional Tutorial Notebook Authors
Other Contributors
(Ordered alphabetically by last name)
