Start here
The course has ten tracks. You don’t need all of them. Pick the path closest to your role; every lesson is also tagged Beginner, Intermediate or Advanced, and Clinical or Technical, so you can adjust as you go.
I'm a clinician, student or health leader
You want to understand how AI tools work, read studies about them critically, and decide whether one is safe and useful in your practice. You don't need to write code, though trying the exercises will deepen your understanding.
- Foundations
- Machine learning
- Generative AI & LLMs
- Evaluation, safety & deployment
- Regulation, ethics & governance
- Applied case studies: diabetes & metabolic health
Focus on lessons tagged Clinical. In Technical lessons, read the explanations and skip the code if you prefer.
I'm a developer, data scientist or researcher
You can code and want to build healthcare AI that holds up outside the lab: with the right data standards, honest evaluation, and an understanding of the clinical and regulatory context.
- Foundations
- Machine learning
- Deep learning & imaging
- Clinical NLP
- Generative AI & LLMs
- Retrieval-augmented generation (RAG)
- Agentic AI
- Evaluation, safety & deployment
- Applied case studies: diabetes & metabolic health
Don't skip Foundations: the data-standards, privacy and diagnostic-statistics lessons are where most clinical AI projects go wrong.
What you need
- For reading: nothing beyond curiosity. Unfamiliar terms are linked to the glossary.
- For exercises: a free Google account to run notebooks in Colab, or a local Python 3 installation.
- No patient data. All exercises use synthetic or openly licensed de-identified datasets.
Educational content only. This material teaches methods; it is not clinical advice. Code, models and examples here are not validated medical devices and must not be used to make decisions about real patients without proper clinical validation, governance and any required regulatory approval. Examples use synthetic or openly licensed de-identified data only. Never paste identifiable patient information into public AI tools. Mentions of models, tools or vendors are for identification, not endorsement.