Title: Dynamic data analysis and data mining for prediction of clinical stability
Authors: Van Loon, Kristien ×
Guiza Grandas, Fabian
Meyfroidt, Geert
Aerts, Jean-Marie
Ramon, Jan
Blockeel, Hendrik
Bruynooghe, Maurice
Van den Berghe, Greet
Berckmans, Daniel #
Issue Date: 2009
Publisher: I O S Press
Host Document: Studies in Health Technology and Informatics vol:150 pages:590-594
Conference: 22th Medical Informatics Europe location:Sarajevo, Bosnia&Herzegovina date:30 august - 2 september 2009
Abstract: This work studies the impact of using dynamic information as features
in a machine learning algorithm for the prediction task of classifying critically ill
patients in two classes according to the time they need to reach a stable state after
coronary bypass surgery: less or more than nine hours. On the basis of five
physiological variables different dynamic features were extracted. These sets of
features served subsequently as inputs for a Gaussian process and the prediction
results were compared with the case where only admission data was used for the
classification. The dynamic features, especially the cepstral coefficients (aROC:
0.749, Brier score: 0.206), resulted in higher performances when compared to
static admission data (aROC: 0.547, Brier score: 0.247). In all cases, the Gaussian
process classifier outperformed logistic regression.
ISSN: 0926-9630
Publication status: published
KU Leuven publication type: IC
Appears in Collections:Division M3-BIORES: Measure, Model & Manage Bioresponses (-)
Informatics Section
Laboratory of Intensive Care Medicine
Unit for Clinical-Translational Research (-)
× corresponding author
# (joint) last author

Files in This Item:
File Description Status SizeFormat
MIE2009_VanLoonK_et_al_revised.pdfarticle Published 126KbAdobe PDFView/Open


All items in Lirias are protected by copyright, with all rights reserved.