Step into the future of Intensive Care innovation by participating in our Hackathon on ICU federated data usage. This dynamic event invites ICU clinicians, engineers, data scientists, and startups from the INDICATE and EIT Health network to co-create a vision for a global, real-time federated data infrastructure. 

Together, we’ll explore how to securely link ICU data across national and EU systems using anonymization techniques and federated learning. While ensuring full compliance with EU regulations and the AI Act. Participants will gain hands-on experience implementing Equitable AI Models Across Diverse European ICU Populations​ and help us build the Value-based proposition of Federated ICU Data Usage, guided by expert mentors. Expect rich cross-pollination of ideas and help shape ethical frameworks for deploying clinical and research algorithms.

Expert speakers

Our first expert speaker Dr. Stefan Buijsman, Associate Professor Responsible AI at TU Delft and member of the INDICATE Ethics Advisory Board, will focus on the first Hackathon topic of Equitable AI Model Performance Across Diverse European ICU Populations.

Our second expert speakers Jan van den Brand, Technical Lead for INDICATE and Hayley Avery, Deputy Director EIT Health Belgium-Netherlands will focus on the second Hackathon topic of Value-based Proposition & Long-term Sustainability.

Paul Hilders, PhD Candidate at Amsterdam UMC, will present the Use Case Virtual digital twin of mechanical ventilation for the Hackathon.

The challenges

Participants will form teams and work intensively over a set period to complete a Problem-Solution Fit Canvas, a tool that facilitates problem analysis and translates it into actionable plans. Through close collaboration, teams will gain a comprehensive understanding of the multi-faceted challenges, which includes technology constraints, regulatory hurdles, resource access, and market demand. This will enable them to conceptualize effective solutions in response to the identified problem. They will also be asked to work on specific outcomes for their chosen challenge:  

• Topic 1 challenge: Equitable AI Model Performance Across Diverse European ICU Populations 
• Topic 2 challenge: Value-based Proposition & Long-term Sustainability 

Jury

The four jury members for the hackathon are:

Dr. Michel van Genderen, is ICU Expert and INDICATE Principal Investigator. Michel is an Intensivist and clinical scientist working at the intersection of critical care, data infrastructure, and responsible artificial intelligence.

Prof. Dr. Maurizio Cecconi, is ICU Expert and member of INDICATE External Expert Advisory Board. Maurizio is an anesthesiologist, intensive care specialist, and Chair of Anesthesia and Intensive Care Units at Humanitas Research Hospital in Milan, Italy.

Dr. Stefan Buijsman, is Equitable model and Ethics Expert and Associate Professor at TU Delft. At TU Delft he focuses on explainable AI as a researcher and as managing director of the TU Delft Digital Ethics Centre, a world-leading centre on the translation of ethical values to concrete design requirements. 

Dr. Paul Elbers, is Mechanical Ventilation (Use Case) Expert and Intensivist and Associate Professor of Intensive Care Medicine at Amsterdam UMC. His research is at the crossroads of artificial intelligence and intensive care medicine. 

Use Case – Virtual digital twin of mechanical ventilation  

This use case is dedicated to developing a virtual digital twin for critically ill patients, a process that involves numerous sequential decisions made every hour. This use case is particularly suited for reinforcement learning, which will be employed to determine the optimal treatment choices tailored to the specific conditions of individual patients at any given moment. 

A primary focus will be optimizing mechanical ventilation settings for treating respiratory failure. Crucial to this use case is the leverage of the MIMIC-EU dataset, as referenced in the INDICATE Use Case. The integration of this diverse and extensive dataset is key in customizing treatment decisions for a wide variety of patient profiles, thereby enhancing the generalizability of the developed models. 

Furthermore, supervised machine learning will be utilized to predict and prevent the need for reintubation and to guide clinicians on data collection for improved predictive accuracy. The outcomes of this work will be disseminated through scientific literature and a pre-print on the knowledge platform, facilitating the rapid exchange of best clinical practices across Europe.

Program

Join us for this collaborative opportunity to discuss how to better implement the ICU federated data infrastructure on Thursday 5th of February from 8:00hrs to 17:00hrs in Amsterdam. Let’s build the future of ICU data. Secure, smart, and shared.

The Hackathon will be a part of the 2026 ESICM Intensive Care Innovation Forum.