MARS

MAchine Reading of patient recordS

Aims

MARS aims to automatically extract information from patient records written in English (in the form of benchmarking datasets) and in Dutch (using data from the university hospitals of KU Leuven) by designing, developing and evaluating advanced language technology for deep semantic processing of the texts which are often morpho-syntactically not well-formed. As a pilot use case, we investigate language technology for profiling and selecting diabetes patients in clinical trials.

Partners

LIIR coordinates the MARS project. The other MARS partners are the Quantitative Lexicology and Variational Linguistics group of KU Leuven (Prof. Ann Bertels and Dr. Kris Heylen), the Centre for IT & IP Law of KU Leuven (Dr. Els Kindt ), and UZ Leuven (Prof. Bart Van den Bosch and Prof. Erwin Bellon).

Results

We have performed first experiments in recognizing temporal relations between events in the natural language texts of clinical records.



Period From 2015-10-01 to 2017-09-30.
Financed by KU Leuven - C22/15/16
Supervised by Marie-Francine Moens
Staff Tuur Leeuwenberg
Contact Tuur Leeuwenberg

Publications

  1. LEEUWENBERG, Artuur & MOENS, Marie-Francine KULeuven-LIIR at SemEval 2016 Task 12: Detecting Narrative Containment in Clinical Records. In Proceedings of the 10th International Workshop on Semantic Evaluation (SemEval 2016). 2016
  2. LEEUWENBERG, Artuur & MOENS, Marie-Francine Structured Learning for Temporal Relation Extraction from Clinical Records. In Proceedings of the Conference of the European Chapter of the Association for Computational Linguistics. 2017
  3. LEEUWENBERG, Tuur & MOENS, Marie-Francine KULeuven-LIIR at SemEval-2017 Task 12: Cross-Domain Temporal Information Extraction from Clinical Records. In Proceedings of the International Workshop on Semantic Evaluation 2017. ACL. 2017


Back to all projects