Exploring the Trade-Off between generalist and specialized Models: A center-based comparative analysis for glioblastoma segmentation
The influence of dataset shift on model performance is evident. Fine-tuned and specialized models, utilizing data from the evaluated center, outperform generalist models, which rely on data from other centers. These approaches could encourage medical centers to develop customized models for their local use, enhancing the accuracy and reliability of glioblastoma segmentation in a context where dataset shift is inevitable.
Authors
- Elies Fuster‐García (ORCID: https://orcid.org/0000-0002-0716-8960)
- Pau Martí (ORCID: https://orcid.org/0000-0002-3222-2658)
- Juan M. García‐Gómez (ORCID: https://orcid.org/0000-0002-3851-1557)
- Pablo Ferri (ORCID: https://orcid.org/0000-0002-3083-1169)
- Víctor Montosa-i-Micó (ORCID: https://orcid.org/0009-0007-2216-9279)
- María Gómez Mahiques (ORCID: https://orcid.org/0009-0003-7138-9340)
- Carles Lopez-Mateu
- F. Javier Gil-Terrón (ORCID: https://orcid.org/0000-0002-3536-6462)
Institutions
- Universitat de les Illes Balears (ES)
- Universitat Politècnica de València (ES)
Publication Details
- Journal
- International Journal of Medical Informatics
- Published
- 2024-08-15
- DOI
- https://doi.org/10.1016/j.ijmedinf.2024.105604
- Citations
- 3
- Primary Topic
- Radiomics and Machine Learning in Medical Imaging
- Type
- article
- Field-Weighted Citation Impact
- 1.21
Funders
- Generalitat Valenciana
- European Regional Development Fund
- Agencia Estatal de Investigación