ESTRO 2024 - Abstract Book

S1812

Clinical - Lung

ESTRO 2024

MedBITR3+ model outperformed the others, achieving the highest score for each evaluated metric (F1- score -F1, Precision -P and Recall R) in 76% of cases.

Conclusion:

The higher performance of the MedBITR3+ relies on its unique approach to pre-training. IN fact the model learns not only from general Italian language but also from the specialized language used in the biomedical field. These results enable to performe biomedical NER on Italian oncology narratives and paves the road for future developments and applications such as integrating such information with other data modalities to perform prognostic and predictive tasks in NSCLC.

Keywords: NSCLC

3123

Mini-Oral

Stereotactic body radiotherapy for extracranial, intracranial or mixed-pattern oligometastatic NSCLC

Jennifer Rogenmoser 1 , Michael C Mayinger 1 , Sebastian M Christ 1 , Eugenia Vlaskou Badra 1 , Maiwand Ahmadsei 1 , Selma Adilovic 1 , Kim Borsky 2 , Simon Frei 2 , Luisa Sabrina Stark Schneebeli 1 , Stephanie Tanadini-Lang 1 , Johannes Kraft 1,3 , Matthias Guckenberger 1 , Nicolaus Andratschke 1 , Jonas Willmann 1 1 University Hospital Zurich and University of Zurich, Department of Radiation Oncology, Zurich, Switzerland. 2 University of Zurich, Faculty of Medicine, Zurich, Switzerland. 3 University Hospital of Wuerzburg, University of Wuerzburg, Department of Radiation Oncology, Wuerzburg, Germany

Purpose/Objective: