From BERT to GPT-4: A systematic review of AI-Driven toxicity extraction and grading in radiation oncology
Current research primarily focused on three basic tasks and three categories of models. Multi-center datasets and secure, lightweight deployment methods are needed before widespread integration into routine radiation oncology practice can be considered.
Authors
- Mark Raymond Waddle (ORCID: https://orcid.org/0000-0002-6017-8850)
- Mi Zhou (ORCID: https://orcid.org/0000-0001-8696-4940)
- Federico Mastroleo (ORCID: https://orcid.org/0000-0001-6580-6767)
- David M. Routman (ORCID: https://orcid.org/0000-0002-2768-003X)
- Satomi Shiraishi (ORCID: https://orcid.org/0000-0002-9522-014X)
- Mohammad Javad Namazi (ORCID: https://orcid.org/0000-0001-5680-3919)
- Anthony Foong (ORCID: https://orcid.org/0009-0004-9467-108X)
- Sarah Peterson (ORCID: https://orcid.org/0009-0004-2011-6366)
- Rebekah Wilson
- Shiv Patel (ORCID: https://orcid.org/0009-0001-6437-3653)
Institutions
- University of Milan (IT)
- Mayo Clinic in Arizona (US)
- European Institute of Oncology (IT)
Publication Details
- Journal
- Technical Innovations & Patient Support in Radiation Oncology
- Published
- 2025-12-19
- DOI
- https://doi.org/10.1016/j.tipsro.2025.100372
- Primary Topic
- Radiomics and Machine Learning in Medical Imaging
- Type
- article
- Field-Weighted Citation Impact
- 0.00