This paper presents an automatic de-identification system for Serbian, based on the adaptation of the existing rule-based named entity recognition system. Built on a finite-state methodology and lexical resources, the system is designed to detect and replace all the explicit personal protected health information present in the medical narrative texts, while still preserving all the relevant medical concepts. The results of a preliminary evaluation demonstrate the usefulness of this method both in preserving patient privacy and the de-identified document interoperability.
Keywords:
named entity recognition / finite-state transducers / rule-based system / de-identification / medical narrative texts / Serbian
TY - JOUR
AU - Jaćimović, Jelena
AU - Krstev, Cvetana
AU - Jelovac, Drago
PY - 2015
UR - https://smile.stomf.bg.ac.rs/handle/123456789/1992
AB - This paper presents an automatic de-identification system for Serbian, based on the adaptation of the existing rule-based named entity recognition system. Built on a finite-state methodology and lexical resources, the system is designed to detect and replace all the explicit personal protected health information present in the medical narrative texts, while still preserving all the relevant medical concepts. The results of a preliminary evaluation demonstrate the usefulness of this method both in preserving patient privacy and the de-identified document interoperability.
PB - Slovensko Drustvo Informatika, Ljubljana
T2 - Informatica - Journal of Computing & Informatics
T1 - A Rule-Based System for Automatic De-identification of Medical Narrative Texts
VL - 39
IS - 1
SP - 43
EP - 51
UR - https://hdl.handle.net/21.15107/rcub_smile_1992
ER -
@article{
author = "Jaćimović, Jelena and Krstev, Cvetana and Jelovac, Drago",
year = "2015",
abstract = "This paper presents an automatic de-identification system for Serbian, based on the adaptation of the existing rule-based named entity recognition system. Built on a finite-state methodology and lexical resources, the system is designed to detect and replace all the explicit personal protected health information present in the medical narrative texts, while still preserving all the relevant medical concepts. The results of a preliminary evaluation demonstrate the usefulness of this method both in preserving patient privacy and the de-identified document interoperability.",
publisher = "Slovensko Drustvo Informatika, Ljubljana",
journal = "Informatica - Journal of Computing & Informatics",
title = "A Rule-Based System for Automatic De-identification of Medical Narrative Texts",
volume = "39",
number = "1",
pages = "43-51",
url = "https://hdl.handle.net/21.15107/rcub_smile_1992"
}
Jaćimović, J., Krstev, C.,& Jelovac, D.. (2015). A Rule-Based System for Automatic De-identification of Medical Narrative Texts. in Informatica - Journal of Computing & Informatics
Slovensko Drustvo Informatika, Ljubljana., 39(1), 43-51.
https://hdl.handle.net/21.15107/rcub_smile_1992
Jaćimović J, Krstev C, Jelovac D. A Rule-Based System for Automatic De-identification of Medical Narrative Texts. in Informatica - Journal of Computing & Informatics. 2015;39(1):43-51.
https://hdl.handle.net/21.15107/rcub_smile_1992 .
Jaćimović, Jelena, Krstev, Cvetana, Jelovac, Drago, "A Rule-Based System for Automatic De-identification of Medical Narrative Texts" in Informatica - Journal of Computing & Informatics, 39, no. 1 (2015):43-51,
https://hdl.handle.net/21.15107/rcub_smile_1992 .
