EIBIR Session

EIBIR 4 - Novel approaches for trustworthy artificial intelligence (AI) in cancer diagnostics and treatment

March 4, 13:00 - 14:30 CET

5 min
Chairperson's introduction
20 min
Developing a validated AI-based decision-making support system for breast cancer
  1. To learn about the latest research in the International Clinical Validation of Radiomics Artificial Intelligence for Breast Cancer Treatment Planning (RadioVal) project.
  2. To explore the application of the Fairness, Universality, Traceability, Usability, Robustness and Explainability (FUTURE)-AI guidelines in developing trustworthy AI tools for breast cancer imaging.
  3. To investigate advancements in AI traceability and explainability to enhance clinical decision support.
  4. To assess current strategies for validating AI systems in breast cancer diagnosis to ensure reliability and clinical effectiveness.
20 min
Learning without sharing. AI collaboration without data exchange through swarm learning
  1. To learn about the latest research in the Open Consortium for Decentralized Medical Artificial Intelligence (ODELIA) project.
  2. To understand the concept of swarm learning and how it enables collaborative AI training without data sharing for MRI-based breast cancer screening as a demonstration case.
  3. To learn how the ODELIA project paves the way for broader adoption of swarm learning in medical AI and beyond.
  4. To recognise the benefits of swarm learning in healthcare innovation to accelerate AI development, enhance diagnostic performance, and foster generalisable solutions.
15 min
Expanding data availability for development and validation of AI tools in breast cancer screening and diagnoses
  1. To learn about the latest research in the Pan-European Breast Image Platform for Advanced AI-based Breast Cancer Screening (BreastSCan) project.
  2. To understand how BreastSCan aligns with EU health data initiatives and Europe's Beating Cancer Plan, fostering real-world AI integration in clinical practice.
  3. To appreciate the impact of cross-border data sharing on advancing cancer diagnostics across Europe to reduce mortality, improve life quality, and advance AI literacy among healthcare professionals and citizens.
  4. To learn how the large-scale, harmonised BreastSCan dataset combining multimodality breast images and clinical data will support the AI tool development for breast cancer screening and diagnosis.
  5. To discuss the priority of future roles of AI tools beyond image quality assessment, breast density classification or lesion detection/characterisation, i.e. personalised breast cancer risk stratification.
15 min
Accelerating trustworthy AI deployment in cancer diagnostics
  1. To learn about the latest research in the Community of Multidisciplinary Professionals Advancing Safe and Successful AI Implementation in Clinical Practice (COMPASS-AI) project.
  2. To recognise the challenges in deploying AI in clinical practice.
  3. To learn about the multidisciplinary community of experts for AI implementation.
  4. To understand the proposed framework for AI deployment guidelines.
15 min
Discussion