Can you tell us more about your role and how you apply data science to your projects?

I work in the cardiology department of the Amsterdam UMC so I apply data science mostly to electrocardiogram and structured data to try and develop models that can simulate and predict patient outcomes. Ultimately our goal is to make digital twins that can combine multiple modalities including imaging and genetics.

Is there a project from this past year that you are most proud of?

We recently demonstrated that we can make generative graph models to build synthetic cohorts that can reproduce really complex statistical properties from the original dataset. Given that we work with multiple centres and highly sensitive data, making potentially a new way of doing multicentre analyses is quite exciting.

What do you like most about being a DSC member?

I like the diversity of the group in terms of subject areas and skillset. When you speak to someone you always get to hear about a novel use case. The centre contains people who can build models and also people who can really implement them.

What is your favourite data science method?

I am a big fan of reinforcement learning. It’s becoming more popular again with agentic systems and fine-tuning large language models and I think it will start to play more of a role with digital twins.

Are you camp Python/R/or something else?

I started using Python back in 2012 so I am more familiar with it than R. I think R has nicer stats packages but Python is the language for building ML models. My first programming language was C++ and I still find uses for it.