PK KUT joins €4M LUMINA Project to Develop AI-based Digital Twins for Lung Cancer Care
Researchers from PK KUT have joined an international consortium awarded €4 million by the European Innovation Council (EIC) to develop a new generation of AI-based clinical decision-support tools. The LUMINA project will create patient-specific digital twins of the lungs to support predictive diagnosis and personalised treatment selection. The funding was granted under the highly competitive EIC Pathfinder Challenges programme. The project, titled LUMINA (Lung Modeling and Intelligence for Advanced Care: A Digital Twin Approach to Lung Cancer Diagnosis and Treatment), integrates generative AI with advanced medical knowledge to translate innovative concepts into real-world clinical impact.
“LUMINA introduces a breakthrough methodology built around an attention-based artificial intelligence architecture designed to integrate longitudinal and multimodal patient data — including CT scans, medical records, laboratory results, and treatment histories. The project also explores innovative solutions such as the BRONCHI DT interactive digital twin for early diagnosis and risk stratification, as well as the ‘Blood-to-Image’ concept, which transforms blood biomarkers into CT-quality visualisations,” says Professor Paweł Pławiak, Dean of the Faculty of Computer Science and Mathematics at PK Krakow University of Technology and coordinator of the university’s contribution to the project.
Researchers from PK KUT play a key role in developing the advanced AI methodologies that support tumour growth modelling and individual treatment response prediction. Their work is essential in merging complex clinical datasets into reliable predictive models that assist doctors in making personalized medical decisions.
Turning Patient Data into Predictive Insights
Lung cancer is often detected at a late stage, making early intervention challenging—especially for patients without traditional risk factors. LUMINA addresses these challenges by modelling how the disease evolves over time and generating predictions that support clinical decision-making. The system is trained on a large dataset of approximately 4,000 patients from Denmark, Italy, Poland, and South Korea, with plans to expand the dataset to more than 10,000 records using synthetic data.
From Research to Future Clinical Impact
The ambition of LUMINA is to establish the scientific foundations for breakthrough innovation in oncology. The goal is to provide an open platform adaptable to other cancer types, offering clinicians and patients proactive, data-driven tools for treatment planning. “The grant we have now received is just the first of several steps towards a solution that – if progress continues successfully – could be implemented and significantly improve lung cancer treatment within 10–15 years,” says Professor Izabela Ewa Nielsen, Project Coordinator and Portfolio Manager, Aalborg University.
Project Facts — LUMINA
Programme & funding: EIC Pathfinder Challenges – challenge area “Generative-AI based agents to revolutionise medical diagnosis and treatment of cancer”. LUMINA is one of seven funded projects in the challenge area, selected from 219 submitted proposals (total allocation: €30 million). LUMINA receives €4 million.
Timeline: 2026–2030 (48-month project).
Coordinator: Aalborg University (Denmark).
Consortium partners:
• Aalborg University (AAU), Denmark – Project Coordinator and Portfolio Manager
• Humanitas Research Hospital / Humanitas AI Center (ICH), Italy
• Politecnico di Milano (POLIMI), Italy
• PK Krakow University of Technology (PK KUT), Poland
• Maria Skłodowska Curie National Research Institute of Oncology (MSCNRIO), Poland
• Region Hovedstaden (REGIONH), Denmark
Disclaimer:
Funded by the European Union. Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or European Innovation Council and SMEs Executive Agency (EISMEA). Neither the European Union nor the granting authority can be held responsible for them.
