AI-BOOST and EUCAIM launch a Generative AI Challenge for Healthcare
The initiative is part of AI-BOOST, an EU-funded open challenge prize programme that supports the development and benchmarking of trustworthy, human-centric AI solutions across Europe.
Designed to run highly replicable AI open innovation competitions across the EU and Associated Countries, AI-BOOST advances major AI fields while fostering collaboration on trustworthy, human-centric solutions to real-world challenges.
Within this framework, Challenge 3, developed jointly with EUCAIM (European Federation for Cancer Images) explores how Generative AI can be used to enhance clinical datasets.
Why this challenge matters?
Medical imaging is a critical source of data for AI models used in diagnosis and prognosis. These models are typically trained on data from clinical trials and research studies, and increasingly from routine healthcare sources known as Real World Data (RWD).
Although RWD can be larger and more representative, it often poses challenges related to completeness, consistency and harmonisation. Datasets may not fully reflect the characteristics of target populations and may lack essential variables or imaging modalities. These limitations can affect model performance and reduce generalisability, particularly for underrepresented patient groups and low-prevalence subgroups.
Generative AI offers a promising way to address these gaps. It can be used to create synthetic data, improve data quality, fill missing values and produce more complete datasets for training and benchmarking. In turn, higher-quality datasets can help reduce bias and support more robust, representative AI models.
This is particularly important when there is imbalance or underrepresentation across demographic characteristics, sex, comorbidities, concurrent conditions, tumour locations and other clinically relevant variables.
Competition format
The AI-BOOST competition offers a two-phase pathway for teams ready to apply generative AI to improve clinical datasets. Entry starts with a simple concept note. Five selected teams will each receive €28 500 and join a five-month development and validation programme, with the chance to compete for a €100 000 final prize.
Open to teams from academia and industry, the AI-BOOST-EUCAIM challenge is a chance to tackle a major barrier in healthcare AI and help make medical imaging and cancer research tools fairer, more reliable and more relevant to real-world care.
The application deadline is 8 September 2026 at 17:00 (CEST).
Join the AI-BOOST Info Webinar on 22 July at 11:00 CEST to learn more about the AI-BOOST Open Innovation Competition and how to apply.
