Vital Prognosis of Patients in Intensive Care Units Using an Ensemble of Bayesian Classifiers

Rosario Delgado; J. David Núñez-González; J. Carlos Yébenes; Ángel Lavado

doi:10.1007/978-3-030-37599-7_51

Vital Prognosis of Patients in Intensive Care Units Using an Ensemble of Bayesian Classifiers

Rosario Delgado, J. David Núñez-González^*, J. Carlos Yébenes, Ángel Lavado

^*Corresponding author for this work

Research output: Chapter in Book › Chapter › Research › peer-review

1 Citation (Scopus)

Abstract

An Ensemble of Bayesian Classifiers (EBC) is constructed to perform vital prognosis of patients in the Intensive Care Units (ICU). The data are scarce and unbalanced, so that the size of the minority class (critically ill patients who die) is very small, and this fact prevents the use of accuracy as a measure of performance in classification; instead we use the Area Under the Precision-Recall curve (AUPR). To address the classification in this setting, we propose the use of an ensemble constructed from five base Bayesian classifiers with the weighted majority vote rule, where the weights are defined from AUPR. We compare this EBC model with the base Bayesian classifiers used to build it, as well as with the ensemble obtained using the mere majority vote criterion, and with some state-of-the-art machine learning supervised classifiers. Our results show that the EBC model outperforms most of the competing classifiers, being only slightly surpassed by Random Forest.

Original language	English
Title of host publication	Machine Learning, Optimization, and Data Science - 5th International Conference, LOD 2019, Proceedings
Editors	Giuseppe Nicosia, Panos Pardalos, Renato Umeton, Giovanni Giuffrida, Vincenzo Sciacca
Pages	619-630
Number of pages	12
DOIs	https://doi.org/10.1007/978-3-030-37599-7_51
Publication status	Published - 2019

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	11943 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Keywords

Area Under the Precision-Recall curve
Bayesian Classifier
Ensemble
ICU
Majority vote
Vital prognosis

Access to Document

10.1007/978-3-030-37599-7_51

Cite this

Delgado, R., Núñez-González, J. D., Yébenes, J. C., & Lavado, Á. (2019). Vital Prognosis of Patients in Intensive Care Units Using an Ensemble of Bayesian Classifiers. In G. Nicosia, P. Pardalos, R. Umeton, G. Giuffrida, & V. Sciacca (Eds.), Machine Learning, Optimization, and Data Science - 5th International Conference, LOD 2019, Proceedings (pp. 619-630). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11943 LNCS). https://doi.org/10.1007/978-3-030-37599-7_51

Delgado, Rosario ; Núñez-González, J. David ; Yébenes, J. Carlos et al. / Vital Prognosis of Patients in Intensive Care Units Using an Ensemble of Bayesian Classifiers. Machine Learning, Optimization, and Data Science - 5th International Conference, LOD 2019, Proceedings. editor / Giuseppe Nicosia ; Panos Pardalos ; Renato Umeton ; Giovanni Giuffrida ; Vincenzo Sciacca. 2019. pp. 619-630 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inbook{80817f4435d54529b0b0c89be11accb8,

title = "Vital Prognosis of Patients in Intensive Care Units Using an Ensemble of Bayesian Classifiers",

abstract = "An Ensemble of Bayesian Classifiers (EBC) is constructed to perform vital prognosis of patients in the Intensive Care Units (ICU). The data are scarce and unbalanced, so that the size of the minority class (critically ill patients who die) is very small, and this fact prevents the use of accuracy as a measure of performance in classification; instead we use the Area Under the Precision-Recall curve (AUPR). To address the classification in this setting, we propose the use of an ensemble constructed from five base Bayesian classifiers with the weighted majority vote rule, where the weights are defined from AUPR. We compare this EBC model with the base Bayesian classifiers used to build it, as well as with the ensemble obtained using the mere majority vote criterion, and with some state-of-the-art machine learning supervised classifiers. Our results show that the EBC model outperforms most of the competing classifiers, being only slightly surpassed by Random Forest.",

keywords = "Area Under the Precision-Recall curve, Bayesian Classifier, Ensemble, ICU, Majority vote, Vital prognosis",

author = "Rosario Delgado and N{\'u}{\~n}ez-Gonz{\'a}lez, {J. David} and Y{\'e}benes, {J. Carlos} and {\'A}ngel Lavado",

note = "Funding Information: Funding Information. R. Delgado and J.D. N{\'u}{\~n}ez-Gonz{\'a}lez are supported by Min-isterio de Econom{\'i}a y Competitividad, Gobierno de Espa{\~n}a, project ref. MTM2015 67802-P. Funding Information: R. Delgado and J.D. N??ez-Gonz?lez are supported by Min-isterio de Econom?a y Competitividad, Gobierno de Espa?a, project ref. MTM2015 67802-P. Publisher Copyright: {\textcopyright} Springer Nature Switzerland AG 2019. Copyright: Copyright 2020 Elsevier B.V., All rights reserved.",

year = "2019",

doi = "10.1007/978-3-030-37599-7_51",

language = "English",

isbn = "9783030375980",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

pages = "619--630",

editor = "Giuseppe Nicosia and Panos Pardalos and Renato Umeton and Giovanni Giuffrida and Vincenzo Sciacca",

booktitle = "Machine Learning, Optimization, and Data Science - 5th International Conference, LOD 2019, Proceedings",

}

Delgado, R, Núñez-González, JD, Yébenes, JC & Lavado, Á 2019, Vital Prognosis of Patients in Intensive Care Units Using an Ensemble of Bayesian Classifiers. in G Nicosia, P Pardalos, R Umeton, G Giuffrida & V Sciacca (eds), Machine Learning, Optimization, and Data Science - 5th International Conference, LOD 2019, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 11943 LNCS, pp. 619-630. https://doi.org/10.1007/978-3-030-37599-7_51

Vital Prognosis of Patients in Intensive Care Units Using an Ensemble of Bayesian Classifiers. / Delgado, Rosario; Núñez-González, J. David; Yébenes, J. Carlos et al.
Machine Learning, Optimization, and Data Science - 5th International Conference, LOD 2019, Proceedings. ed. / Giuseppe Nicosia; Panos Pardalos; Renato Umeton; Giovanni Giuffrida; Vincenzo Sciacca. 2019. p. 619-630 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11943 LNCS).

Research output: Chapter in Book › Chapter › Research › peer-review

TY - CHAP

T1 - Vital Prognosis of Patients in Intensive Care Units Using an Ensemble of Bayesian Classifiers

AU - Delgado, Rosario

AU - Núñez-González, J. David

AU - Yébenes, J. Carlos

AU - Lavado, Ángel

N1 - Funding Information: Funding Information. R. Delgado and J.D. Núñez-González are supported by Min-isterio de Economía y Competitividad, Gobierno de España, project ref. MTM2015 67802-P. Funding Information: R. Delgado and J.D. N??ez-Gonz?lez are supported by Min-isterio de Econom?a y Competitividad, Gobierno de Espa?a, project ref. MTM2015 67802-P. Publisher Copyright: © Springer Nature Switzerland AG 2019. Copyright: Copyright 2020 Elsevier B.V., All rights reserved.

PY - 2019

Y1 - 2019

N2 - An Ensemble of Bayesian Classifiers (EBC) is constructed to perform vital prognosis of patients in the Intensive Care Units (ICU). The data are scarce and unbalanced, so that the size of the minority class (critically ill patients who die) is very small, and this fact prevents the use of accuracy as a measure of performance in classification; instead we use the Area Under the Precision-Recall curve (AUPR). To address the classification in this setting, we propose the use of an ensemble constructed from five base Bayesian classifiers with the weighted majority vote rule, where the weights are defined from AUPR. We compare this EBC model with the base Bayesian classifiers used to build it, as well as with the ensemble obtained using the mere majority vote criterion, and with some state-of-the-art machine learning supervised classifiers. Our results show that the EBC model outperforms most of the competing classifiers, being only slightly surpassed by Random Forest.

AB - An Ensemble of Bayesian Classifiers (EBC) is constructed to perform vital prognosis of patients in the Intensive Care Units (ICU). The data are scarce and unbalanced, so that the size of the minority class (critically ill patients who die) is very small, and this fact prevents the use of accuracy as a measure of performance in classification; instead we use the Area Under the Precision-Recall curve (AUPR). To address the classification in this setting, we propose the use of an ensemble constructed from five base Bayesian classifiers with the weighted majority vote rule, where the weights are defined from AUPR. We compare this EBC model with the base Bayesian classifiers used to build it, as well as with the ensemble obtained using the mere majority vote criterion, and with some state-of-the-art machine learning supervised classifiers. Our results show that the EBC model outperforms most of the competing classifiers, being only slightly surpassed by Random Forest.

KW - Area Under the Precision-Recall curve

KW - Bayesian Classifier

KW - Ensemble

KW - ICU

KW - Majority vote

KW - Vital prognosis

UR - http://www.scopus.com/inward/record.url?scp=85078456583&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-37599-7_51

DO - 10.1007/978-3-030-37599-7_51

M3 - Chapter

AN - SCOPUS:85078456583

SN - 9783030375980

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 619

EP - 630

BT - Machine Learning, Optimization, and Data Science - 5th International Conference, LOD 2019, Proceedings

A2 - Nicosia, Giuseppe

A2 - Pardalos, Panos

A2 - Umeton, Renato

A2 - Giuffrida, Giovanni

A2 - Sciacca, Vincenzo

ER -

Delgado R, Núñez-González JD, Yébenes JC, Lavado Á. Vital Prognosis of Patients in Intensive Care Units Using an Ensemble of Bayesian Classifiers. In Nicosia G, Pardalos P, Umeton R, Giuffrida G, Sciacca V, editors, Machine Learning, Optimization, and Data Science - 5th International Conference, LOD 2019, Proceedings. 2019. p. 619-630. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-030-37599-7_51

Vital Prognosis of Patients in Intensive Care Units Using an Ensemble of Bayesian Classifiers

Abstract

Publication series

Keywords

Access to Document

Other files and links

Fingerprint

Cite this