Rein Houthooft1, Joeri Ruyssinck2, Joachim van der Herten2, Sean Stijven2, Ivo Couckuyt2, Bram Gadeyne3, Femke Ongenae2, Kirsten Colpaert4, Johan Decruyenaere5, Tom Dhaene2, Filip De Turck2. 1. Department of Information Technology (INTEC), Ghent University - iMinds, Gaston Crommenlaan 8, B-9050 Ghent, Belgium. Electronic address: rein.houthooft@intec.ugent.be. 2. Department of Information Technology (INTEC), Ghent University - iMinds, Gaston Crommenlaan 8, B-9050 Ghent, Belgium. 3. Department of Information Technology (INTEC), Ghent University - iMinds, Gaston Crommenlaan 8, B-9050 Ghent, Belgium; Department of Intensive Care Medicine, Ghent University Hospital, De Pintelaan 185, 2 K12 IC, B-9000 Ghent, Belgium. 4. Department of Intensive Care Medicine, Ghent University Hospital, De Pintelaan 185, 2 K12 IC, B-9000 Ghent, Belgium. 5. Department of Intensive Care Medicine, Ghent University Hospital, De Pintelaan 185, 2 K12 IC, B-9000 Ghent, Belgium; Department of Internal Medicine, Ghent University, De Pintelaan 185, B-9000 Ghent, Belgium.
Abstract
INTRODUCTION: The length of stay of critically ill patients in the intensive care unit (ICU) is an indication of patient ICU resource usage and varies considerably. Planning of postoperative ICU admissions is important as ICUs often have no nonoccupied beds available. PROBLEM STATEMENT: Estimation of the ICU bed availability for the next coming days is entirely based on clinical judgement by intensivists and therefore too inaccurate. For this reason, predictive models have much potential for improving planning for ICU patient admission. OBJECTIVE: Our goal is to develop and optimize models for patient survival and ICU length of stay (LOS) based on monitored ICU patient data. Furthermore, these models are compared on their use of sequential organ failure (SOFA) scores as well as underlying raw data as input features. METHODOLOGY: Different machine learning techniques are trained, using a 14,480 patient dataset, both on SOFA scores as well as their underlying raw data values from the first five days after admission, in order to predict (i) the patient LOS, and (ii) the patient mortality. Furthermore, to help physicians in assessing the prediction credibility, a probabilistic model is tailored to the output of our best-performing model, assigning a belief to each patient status prediction. A two-by-two grid is built, using the classification outputs of the mortality and prolonged stay predictors to improve the patient LOS regression models. RESULTS: For predicting patient mortality and a prolonged stay, the best performing model is a support vector machine (SVM) with GA,D=65.9% (area under the curve (AUC) of 0.77) and GS,L=73.2% (AUC of 0.82). In terms of LOS regression, the best performing model is support vector regression, achieving a mean absolute error of 1.79 days and a median absolute error of 1.22 days for those patients surviving a nonprolonged stay. CONCLUSION: Using a classification grid based on the predicted patient mortality and prolonged stay, allows more accurate modeling of the patient LOS. The detailed models allow to support the decisions made by physicians in an ICU setting.
INTRODUCTION: The length of stay of critically illpatients in the intensive care unit (ICU) is an indication of patient ICU resource usage and varies considerably. Planning of postoperative ICU admissions is important as ICUs often have no nonoccupied beds available. PROBLEM STATEMENT: Estimation of the ICU bed availability for the next coming days is entirely based on clinical judgement by intensivists and therefore too inaccurate. For this reason, predictive models have much potential for improving planning for ICU patient admission. OBJECTIVE: Our goal is to develop and optimize models for patient survival and ICU length of stay (LOS) based on monitored ICU patient data. Furthermore, these models are compared on their use of sequential organ failure (SOFA) scores as well as underlying raw data as input features. METHODOLOGY: Different machine learning techniques are trained, using a 14,480 patient dataset, both on SOFA scores as well as their underlying raw data values from the first five days after admission, in order to predict (i) the patient LOS, and (ii) the patient mortality. Furthermore, to help physicians in assessing the prediction credibility, a probabilistic model is tailored to the output of our best-performing model, assigning a belief to each patient status prediction. A two-by-two grid is built, using the classification outputs of the mortality and prolonged stay predictors to improve the patient LOS regression models. RESULTS: For predicting patient mortality and a prolonged stay, the best performing model is a support vector machine (SVM) with GA,D=65.9% (area under the curve (AUC) of 0.77) and GS,L=73.2% (AUC of 0.82). In terms of LOS regression, the best performing model is support vector regression, achieving a mean absolute error of 1.79 days and a median absolute error of 1.22 days for those patients surviving a nonprolonged stay. CONCLUSION: Using a classification grid based on the predicted patient mortality and prolonged stay, allows more accurate modeling of the patient LOS. The detailed models allow to support the decisions made by physicians in an ICU setting.
Authors: Kan Wang; Li Zhao Yan; Wang Zi Li; Chen Jiang; Ni Ni Wang; Qiang Zheng; Nian Guo Dong; Jia Wei Shi Journal: Front Cardiovasc Med Date: 2022-06-21
Authors: Joeri Ruyssinck; Joachim van der Herten; Rein Houthooft; Femke Ongenae; Ivo Couckuyt; Bram Gadeyne; Kirsten Colpaert; Johan Decruyenaere; Filip De Turck; Tom Dhaene Journal: Comput Math Methods Med Date: 2016-10-13 Impact factor: 2.238
Authors: José Carlos R Alcantud; Gonzalo Varela; Beatriz Santos-Buitrago; Gustavo Santos-García; Marcelo F Jiménez Journal: PLoS One Date: 2019-06-19 Impact factor: 3.240
Authors: Gabriel Fernando Todeschi Variane; João Paulo Vasques Camargo; Daniela Pereira Rodrigues; Maurício Magalhães; Marcelo Jenné Mimica Journal: Front Pediatr Date: 2022-03-23 Impact factor: 3.418
Authors: Igor Tona Peres; Silvio Hamacher; Fernando Luiz Cyrino Oliveira; Fernando Augusto Bozza; Jorge Ibrain Figueira Salluh Journal: Rev Bras Ter Intensiva Date: 2021 Apr-Jun