Post-processing steps improve generalisability and robustness of an MRI-based radiogenomic model for human papillomavirus status prediction in oropharyngeal cancer
Question How do post-processing steps impact the generalisability of MRI-based radiogenomic prediction models? Findings Applying post-processing steps, i.e., data harmonisation, identification of stable radiomic features, and removal of correlated features, before model development can improve model robustness and generalisability. Clinical relevance Post-processing steps in MRI radiogenomic model generation lead to reliable non-invasive diagnostic tools for personalised cancer treatment strategies.
Authors
- Georgios Agrotis (ORCID: https://orcid.org/0000-0002-3752-1315)
- Bas Jasperse (ORCID: https://orcid.org/0000-0003-1577-8299)
- Paula D. Bos (ORCID: https://orcid.org/0000-0001-8035-5888)
- Michiel W. M. van den Brekel (ORCID: https://orcid.org/0000-0002-6338-6743)
- Roland M. Martens (ORCID: https://orcid.org/0000-0002-8297-6802)
- L.H.E. Karssemakers (ORCID: https://orcid.org/0000-0002-0257-8709)
- Jonas A. Castelijns (ORCID: https://orcid.org/0000-0001-5105-4023)
- C. René Leemans (ORCID: https://orcid.org/0000-0001-7887-083X)
- Hedda J. van der Hulst (ORCID: https://orcid.org/0000-0002-6932-9838)
- Conchita Vens (ORCID: https://orcid.org/0000-0002-8608-7236)
- Ruud H. Brakenhoff (ORCID: https://orcid.org/0000-0003-3251-5378)
- Zuhir Bodalal (ORCID: https://orcid.org/0000-0002-2617-8128)
- Abrahim Al‐Mamgani (ORCID: https://orcid.org/0000-0001-5240-6510)
- Pim de Graaf (ORCID: https://orcid.org/0000-0003-1938-0747)
- Milad Ahmadian (ORCID: https://orcid.org/0000-0003-3165-9409)
- Geerard L. Beets (ORCID: https://orcid.org/0000-0002-1671-9912)
Institutions
- University of Southern Denmark (DK)
- Dutch Cancer Society (NL)
- The Netherlands Cancer Institute (NL)
- Maastricht University (NL)
- University Hospital of Larissa (GR)
- Netherlands Cancer Institute-Antoni van Leeuwenhoek Hospital (NL)
- Amsterdam University Medical Centers (NL)
- Cancer Center Amsterdam (NL)
- Amsterdam University of Applied Sciences (NL)
- University of Glasgow (GB)
- Vrije Universiteit Amsterdam (NL)
- University of Amsterdam (NL)
Publication Details
- Journal
- European Radiology
- Published
- 2025-06-06
- DOI
- https://doi.org/10.1007/s00330-025-11709-8
- Citations
- 3
- Primary Topic
- Radiomics and Machine Learning in Medical Imaging
- Type
- article
- Field-Weighted Citation Impact
- 4.08
Funders
- Hanarth Fonds