Automating Performance Status Annotation in Oncology Using Llama-3
This work explores the automated extraction of medical information from Dutch clinical notes using Llama-3 and a limited amount of annotations. We compared zero-, one- and few-shot learning for the extraction of performance status of patients with palliative esophagogastric cancer few-shot learning and one-shot with ACSESS-selected examples, showed the best performance. Future work shall focus on improving the model's precision.
Authors
- Rob H.A. Verhoeven (ORCID: https://orcid.org/0000-0001-9678-1611)
- Linde M. Veen (ORCID: https://orcid.org/0000-0003-0416-4013)
- Gijs Geleijnse (ORCID: https://orcid.org/0000-0002-0947-8970)
- Sebastiaan Siegerink (ORCID: https://orcid.org/0009-0000-3692-844X)
- Hanneke W. M. van Laarhoven (ORCID: https://orcid.org/0000-0003-3546-9709)
- Irene Cara (ORCID: https://orcid.org/0000-0003-3396-2532)
- Nynke van ’t Hof
Institutions
- Netherlands Comprehensive Cancer Organisation (NL)
- Amsterdam University Medical Centers (NL)
- Vrije Universiteit Amsterdam (NL)
Publication Details
- Journal
- Studies in health technology and informatics
- Published
- 2025-05-15
- DOI
- https://doi.org/10.3233/shti250483
- Primary Topic
- Radiomics and Machine Learning in Medical Imaging
- Type
- article
- Field-Weighted Citation Impact
- 0.00