ECIR 2023
SimpleRad: patient-friendly Dutch radiology reports
K. Dercksen, A.P. de Vries, B. van Ginneken
SimpleRad is a tool for entity linking, summarization and finding prevalence estimation in Dutch radiology reports.
SIGIR 2020
REL: an entity linker standing on the shoulders of giants
J.M. van Hulst, F. Hasibi, K. Dercksen, K. Balog, A. P. de Vries
REL (Radboud Entity Linker) is a modular entity linking framework based on state-of-the-art neural components. Code is available for local deployment. A web API is available as well.
CLEF 2020
Named entity recognition and linking on historical newspapers: UvA. ILPS & REL at CLEF HIPE 2020
V. Provatorova, S. Vakulenko, E. Kanoulas, K. Dercksen, J.M. van Hulst
Paper describing our submission to the CLEF HIPE 2020 shared task on identifying named entities in multi-lingual historical newspapers in French, German and English.
SIIRH 2020
First steps towards patient-friendly presentation of Dutch radiology reports
K. Dercksen, A.P. de Vries
Paper describing our initial efforts towards patient-friendly Dutch radiology reports. Primarily on entity linking annotation and future work.
arXiv
Exploring task-based query expansion at the TREC-COVID track
T. Schoegje, C. Kamphuis, K. Dercksen, D. Hiemstra, T. Pieters, A. P. de Vries
Exploring how to generate effective queries based on search tasks for the TREC-COVID track.
IWBI 2020
Robust multi-vendor breast region segmentation using deep learning
K. Dercksen, M. Kallenberg, J. Kroes
This paper investigates a robust multi-vendor breast region segmentation system for full field digital mammograms and digital breast tomography using a U-Net neural network.
MIDL 2019
Dealing with label scarcity in computational pathology: a use case in prostate cancer classification
K. Dercksen, W. Bulten, G. Litjens
This paper investigates the performance of unsupervised and supervised deep learning methods for prostate cancer classification when few labelled data are available.