Elyne Scheurwegs




Homepage


PhD-student at Computational Linguistics and Psycholinguistics Research Center (CLiPS) and Advanced Database Research and Modelling (ADReM), University of Antwerp
External researcher at Antwerp University Hospital

University of Antwerp
Dept. of Mathematics and Computer Science

Office G.322
Middelheimlaan 1
2020 Antwerpen
Belgium

E-mail: elyne.scheurwegs@uantwerpen.be

About me

I currently work as a PhD student at the University of Antwerp, under the supervision of prof. Bart Goethals and prof. Walter Daelemans. Previously, I obtained a MSc in Industrial Sciences: Electronics-ICT Engineering at the Artesis College in Antwerp.

Research project

My current PhD project is orientated towards predicting clinical codes (e.g. ICD-9-CM codes) for patient stays. This involves managing data from both structured and unstructured data sources in a real-life medical environment. The main issues tackled in this project are efficiently handling a large multi-class problem, managing data with a large amount of noise, retrieving complementary information in different sources (with a focus on representing and processing free text) and machine learning with a limited number of annotated examples.

Publications

2015

Data integration of structured and unstructured sources for assigning clinical codes to patient stays
Elyne Scheurwegs, Kim Luyckx, Léon Luyten, Walter Daelemans, Tim Van den Bulcke
Journal of the American Medical Informatics Association (JAMIA)


2013

De-Identification of Clinical Free Text in Dutch with Limited Training Data: A Case Study.
Elyne Scheurwegs, Kim Luyckx, Filip Van der Schueren & Tim Van den Bulcke
In Proceedings of the Workshop on NLP for Medicine and Biology associated with RANLP (pp. 18-23).

Talks & posters

2015

  • [Talk] Biomina Research Day 2015; 9 December 2015; Integration of structured and unstructured data sources for predicting clinical codes (work with Kim Luyckx, Léon Luyten, Walter Daelemans and Tim Van den Bulcke)
  • [Poster Presentation] Benelux Bioinformatics Conference 2015 (BBC)7-8 December 2015 Integration of structured and unstructured data sources for predicting clinical codes (work with Kim Luyckx, Léon Luyten, Walter Daelemans and Tim Van den Bulcke)
  • [Talk] ATILA 2015; 15-16 October 2015; Representing patient stays with word2vec for predicting clinical codes (work with Kim Luyckx, Léon Luyten, Bart Goethals and Walter Daelemans)
  • [Poster Presentation] 25th Meeting of Computational Linguistics in The Netherlands (CLIN25); 5-6 February 2015; Comparing NLP techniques to represent unstructured data sources for the prediction of clinical codes (work with Kim Luyckx, Léon Luyten, Walter Daelemans and Tim Van den Bulcke)

2014

  • [Poster Presentation] Biomina Research Day 2014; 20 February 2015; Automated Clinical Coding: A problem analysis (work with Kim Luyckx, Léon Luyten, Walter Daelemans and Tim Van den Bulcke)