Impact of harmonization and oversampling methods on radiomics analysis of multi-center imbalanced datasets: application to PET-based prediction of lung cancer subtypes
Applying harmonization and oversampling methods in multi-center imbalanced datasets can improve NSCLC-subtype prediction, but the effect varies widely across classifiers. We have created open-source comparisons of harmonization and oversampling on different classifiers for comprehensive evaluations in different studies.
Authors
- Isaac Shiri (ORCID: https://orcid.org/0000-0002-5735-0736)
- Habib Zaidi (ORCID: https://orcid.org/0000-0001-7559-5297)
- Arman Rahmim (ORCID: https://orcid.org/0000-0002-9980-2403)
- Lijun Lu (ORCID: https://orcid.org/0000-0002-7001-4892)
- Wentao Zhu (ORCID: https://orcid.org/0000-0001-9290-1778)
- Dongyang Du
- Fereshteh Yousefirizi (ORCID: https://orcid.org/0000-0001-5261-6163)
- Jieqin Lv
- Mohammad R. Salmanpour (ORCID: https://orcid.org/0000-0002-9515-789X)
- Huiqin Wu
Institutions
- University of British Columbia (CA)
- Inner Mongolia University (CN)
- Key Laboratory of Guangdong Province (CN)
- University Hospital of Geneva (CH)
- Zhejiang Lab (CN)
- Southern Medical University (CN)
Publication Details
- Journal
- EJNMMI Physics
- Published
- 2025-04-07
- DOI
- https://doi.org/10.1186/s40658-025-00750-7
- Citations
- 2
- Primary Topic
- Radiomics and Machine Learning in Medical Imaging
- Type
- article
- Field-Weighted Citation Impact
- 2.72
Funders
- National Natural Science Foundation of China
- Natural Science Foundation of Inner Mongolia
- Natural Sciences and Engineering Research Council of Canada
- Science and Technology Planning Project of Guangdong Province