Supplementary material for "Reshaping free-text radiology notes into structured reports with generative question answering transformers" paper
-
Pre-processing phase carried out on radiology reports and the related human-expert annotations: colab notebook
-
SR registry filling pipeline: SCRIPTS
-
Comparative analysis of models performances and human-expert evaluation analysis: colab notebook
@article{BERGOMI2024102924,
title = {Reshaping free-text radiology notes into structured reports with generative question answering transformers},
journal = {Artificial Intelligence in Medicine},
pages = {102924},
year = {2024},
issn = {0933-3657},
doi = {https://doi.org/10.1016/j.artmed.2024.102924},
url = {https://www.sciencedirect.com/science/article/pii/S0933365724001660},
author = {Laura Bergomi and Tommaso M. Buonocore and Paolo Antonazzo and Lorenzo Alberghi and Riccardo Bellazzi and Lorenzo Preda and Chandra Bortolotto and Enea Parimbelli},
keywords = {Natural language processing, Clinical text, Generative artificial intelligence, Radiology, Lymphoma, Biomedical information extraction}
}