Breast cancer remains a major global heath challenge. Despite progress in screening and treatment, many patients still experience poor outcomes, highlighting the need for more precise and cost-effective diagnostic tools. The CARE-B project aims to advance precision diagnostics in breast cancer by applying AI to model registry data, histopathology images, and molecular profiles from cohorts in Sweden, Denmark, and Scotland.

By building a large, multimodal data resource and together with scalable AI models, the project seeks to improve patient stratification, prediction of patient outcomes and enable characterisation of  intra-tumor heterogeneity. CARE-B focus on collaboration across partners, support for early-career researchers, and on creating a foundation for AI-based precision medicine and epidemiological research at the forefront.

The goal of our project is to bring together AI methods with large scale epidemiological studies to further advance precision diagnostics and ultimately enhance patient outcomes in breast cancer care.

• Creating a large, multimodal database (up to 30,000 patients) combining clinical data, whole slide histopathology images, and molecular profiles from Sweden, Denmark, and Scotland.

• Developing scalable AI models for:
  • Deep phenotyping of breast cancer subtypes using routine H&E slides.
  • Characterizing intra-tumor heterogeneity (ITH).
  • Predicting patient outcomes including both treatment responses and prognostic stratification.

• Fostering collaboration and supporting junior researchers, while pushing the boundaries of registry-based epidemiological research.

Karolinska Institutet

Karolinska Institutet coordinates the consortium, leading development of AI-based precision pathology and providing core infrastructure, registry linkage, and epidemiology expertise for clinical translation.

Researchers: Senior lecturer and Docent Mattias Rantalainen, Prof. Johan Hartman, Dr. Bojing Liu

The university of Edinburgh

The University of Edinburgh contributes a large real‑world clinical breast cancer cohort and leads health‑economics, real-world evidence, and model validation to support translation into healthcare systems.

Researchers: Prof. Peter Hall, Dr. Karen Taylor, Dr. Azadeh Abravan

University of Copenhagen / Herlev and Gentofte Hospital (HGH)

Herlev and Gentofte Hospital  provides nationwide Danish breast cancer registry and pathology data and leads clinical validation to ensure cross-country generalisability of AI-based markers.

Researcher: Dr. Anne-Vibeke Laenkholm

Lund University

Lund University provides the SCAN‑B cohort, the world's largest population-based breast cancer RNA-seq resource, and leads development and validation of RNA-based biomarkers.

Researcher: Associate Prof. Johan Vallon-Christersson

Contact

PI and coordinator: Mattias Rantalainen,  mattias.rantalainen@ki.se

Project manager: Anne-May Österholm, anne-may.osterholm@ki.se