External Validation of a CT-Based Radiogenomics Model for the Detection of EGFR Mutation in NSCLC and the Impact of Prevalence in Model Building by Using Synthetic Minority Over Sampling (SMOTE): Lessons Learned
External validation failed to reproduce prior radiogenomics model performance, while clinical variables alone retained strong predictive value. SMOTE-based oversampling did not improve diagnostic accuracy, suggesting that, in EGFR prediction, radiomics may offer limited value beyond clinical data. Emphasis on robust external validation and data-sharing is essential for future clinical implementation of radiogenomic models.
Authors
- Andrés Kohan (ORCID: https://orcid.org/0000-0003-3592-3000)
- Patrick Veit‐Haibach (ORCID: https://orcid.org/0000-0002-8607-2667)
- Andrew Hope (ORCID: https://orcid.org/0000-0001-6793-2863)
- R. Kulanthaivelu
- Ricarda Hinzpeter (ORCID: https://orcid.org/0000-0002-4707-3042)
- Ur Metser (ORCID: https://orcid.org/0000-0003-3726-8994)
- Lisa M. Avery (ORCID: https://orcid.org/0000-0002-8431-5143)
- Claudia Ortega (ORCID: https://orcid.org/0000-0003-3357-1849)
- S.A. Mirshahvalad
Institutions
- University Health Network (CA)
- Princess Margaret Cancer Centre (CA)
- University Hospital of Zurich (CH)
Publication Details
- Journal
- Academic Radiology
- Published
- 2025-05-15
- DOI
- https://doi.org/10.1016/j.acra.2025.04.064
- Citations
- 2
- Primary Topic
- Radiomics and Machine Learning in Medical Imaging
- Type
- article
- Field-Weighted Citation Impact
- 2.72