Published January 14, 2025 | Version v1
Journal Open

Leveraging Knowledge for Explainable AI in Personalized Cancer Treatment: Challenges and Future Directions

Creators

  • 1. International Centre for Cancer Vaccine Science, University of Gdansk, Gdansk, Poland
  • 2. ROR icon University of Rome Tor Vergata
  • 3. Urology Department, Western General Hospital, NHS Lothian, Edinburgh, UK
  • 4. PredictBy Research and Consulting SLU
  • 5. Caretronic d.o.o.Štirnova 8, 4000 Kranj, Slovenia
  • 6. DSTECH S.r.l. Via Salaria 719, 0028 – Rome, Italy
  • 7. CEA, Centre National de Recherche en Génomique Humaine, Université Paris-Saclay, Evry, France
  • 8. Centre Hospitalier Universitaire de Grenoble (CHU), Grenoble, France
  • 9. LASIGE, Faculdade de Ciências, Universidade de Lisboa, Lisbon, Portugal
  • 10. Université Grenoble Alpes, IRIG, Laboratoire Biosciences et Bioingénierie pour la Santé, UA2 INSERM-CEA-UGA, 38000 Grenoble, France
  • 11. Personal Genomics Srl, Via Roveggia 43B, 3726, Verona, Italy
  • 12. Eurecat, Centre Tecnològic de Catalunya, Barcelona, 08005, Spain
  • 13. Fondazione IRCCS Istituto Nazionale dei Tumori (INT), Via Giacomo Venezian 1, 2023, Milan, Itlay
  • 14. University of Vienna, Department of Innovation and Digitalisation in Law, Vienna, Austria
  • 15. Laboratory of Biology, Medical School, National and Kapodistrian University of Athens, Athens, Greece
  • 16. School of Medicine, University of St Andrews, St Andrews, UK
  • 17. Health Policy Institute Civil Non for Profit Company, HPI, Greece
  • 18. School of Informatics, University of Edinburgh, Edinburgh, UK
  • 19. Institute of Genetics and Cancer, University of Edinburgh, Edinburgh, UK
  • 20. National Technical University of Ukraine Igor Sikorsky Kyiv Polytechnic Institute, 37 Prospect Peremohy, Kyiv, Ukraine
  • 21. EURICE, Heinrich-Hertz-Allee 1, 66386 St. Ingbert, Germany
  • 22. Division of Translational Cancer Research, Department of Laboratory Medicine, Lund University, Lund, Sweden
  • 23. PredictBy Research and Consulting SLU, Universidad Oberta Catalunya, Barcelona, Spain
  • 24. Institute for Biocomputation and Physics of Complex Systems and Department of Theoretical Physics, University of Zaragoza, Zaragoza 50018, Spain

Description

The rapid expansion of disease-specific knowledge poses challenges for clinicians to remain up-to-date and effectively leverage emerging information into clinical practice. Addressing this challenge requires implementation of an informed, evidence-based decision-support system continuously maintained and updated to align with the latest clinical guidelines supporting personalized decision-making in real-time. Knowledge graphs have emerged as data models, integrating multi-modal (multi-omics) data sources and representing them as interconnected networks. These graphs offer a framework which easily streamlines the iterative learning process for AI models facilitating transparency and explainability of complex interactions and interdependencies. In this paper, we discuss the major components, challenges and future directions of an explainable AI-supported system in the biomedical domain emphasizing its transformative potential for the healthcare system. We discuss the concept of modular multi-tiered AI-supported system architecture, data privacy and security issues along with existing technical barriers that should be overcome to develop an AI-driven decision support system capable of handling the complexity of multi-modal data, being adaptable to new knowledge and constantly evolving treatment guidelines. Finally, we showcase ongoing efforts by the Knowledge at the Tips of your Fingers (KATY) consortium to implement AI-assisted strategies for enhancing the contribution of -omics studies in personalized medicine.

Files

Manuscript.pdf

Files (671.4 kB)

Name Size Download all
md5:107f18a42fc235707830a743fa9b03f3
671.4 kB Preview Download

Additional details

Funding

European Commission
KATY - Knowledge At the Tip of Your fingers: Clinical Knowledge for Humanity 101017453

Dates

Created
2025-01-14