Tampere University of Technology

TUTCRIS Research Portal

Sepsis Prediction in Intensive Care Unit Using Ensemble of XGboost Models

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Standard

Sepsis Prediction in Intensive Care Unit Using Ensemble of XGboost Models. / Zabihi, Morteza; Kiranyaz, Serkan; Gabbouj, Moncef.

2019 Computing in Cardiology, CinC 2019. IEEE Computer Society, 2019. 9005564 (Computing in Cardiology; Vol. 2019-September).

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Harvard

Zabihi, M, Kiranyaz, S & Gabbouj, M 2019, Sepsis Prediction in Intensive Care Unit Using Ensemble of XGboost Models. in 2019 Computing in Cardiology, CinC 2019., 9005564, Computing in Cardiology, vol. 2019-September, IEEE Computer Society, 2019 Computing in Cardiology, CinC 2019, Singapore, Singapore, 8/09/19. https://doi.org/10.23919/CinC49843.2019.9005564

APA

Zabihi, M., Kiranyaz, S., & Gabbouj, M. (2019). Sepsis Prediction in Intensive Care Unit Using Ensemble of XGboost Models. In 2019 Computing in Cardiology, CinC 2019 [9005564] (Computing in Cardiology; Vol. 2019-September). IEEE Computer Society. https://doi.org/10.23919/CinC49843.2019.9005564

Vancouver

Zabihi M, Kiranyaz S, Gabbouj M. Sepsis Prediction in Intensive Care Unit Using Ensemble of XGboost Models. In 2019 Computing in Cardiology, CinC 2019. IEEE Computer Society. 2019. 9005564. (Computing in Cardiology). https://doi.org/10.23919/CinC49843.2019.9005564

Author

Zabihi, Morteza ; Kiranyaz, Serkan ; Gabbouj, Moncef. / Sepsis Prediction in Intensive Care Unit Using Ensemble of XGboost Models. 2019 Computing in Cardiology, CinC 2019. IEEE Computer Society, 2019. (Computing in Cardiology).

Bibtex - Download

@inproceedings{0744fb7220b24e3eb831d7beafc97bd4,
title = "Sepsis Prediction in Intensive Care Unit Using Ensemble of XGboost Models",
abstract = "Sepsis is caused by the dysregulated host response to infection and potentially is the main cause of 6 million death annually. It is a highly dynamic syndrome and therefore the early prediction of sepsis plays a key role in reducing its high associated mortality. However, this is a challenging task because there is no specific and accurate test or scoring system to perform early prediction. In this paper, we present a systematic approach for sepsis prediction. We also propose a new set of features to model the missingness in clinical data. The pipeline of the proposed method comprises three major components: feature extraction, feature selection, and classification. In total, 407 features are extracted from the clinical data. Then, five different sets of features are selected using a wrapper feature selection algorithm based on XGboost. The selected features are extracted from both valid and missing clinical data. Afterwards, an ensemble model consists of five XGboost models is used for sepsis prediction. The proposed algorithm is ranked officially as third place in the PhysioNet/Computing in Cardiology Challenge 2019 with an overall utility score of 0.339 on the unseen test dataset (our team name: Separatrix).",
author = "Morteza Zabihi and Serkan Kiranyaz and Moncef Gabbouj",
year = "2019",
month = "9",
day = "1",
doi = "10.23919/CinC49843.2019.9005564",
language = "English",
series = "Computing in Cardiology",
publisher = "IEEE Computer Society",
booktitle = "2019 Computing in Cardiology, CinC 2019",
address = "United States",

}

RIS (suitable for import to EndNote) - Download

TY - GEN

T1 - Sepsis Prediction in Intensive Care Unit Using Ensemble of XGboost Models

AU - Zabihi, Morteza

AU - Kiranyaz, Serkan

AU - Gabbouj, Moncef

PY - 2019/9/1

Y1 - 2019/9/1

N2 - Sepsis is caused by the dysregulated host response to infection and potentially is the main cause of 6 million death annually. It is a highly dynamic syndrome and therefore the early prediction of sepsis plays a key role in reducing its high associated mortality. However, this is a challenging task because there is no specific and accurate test or scoring system to perform early prediction. In this paper, we present a systematic approach for sepsis prediction. We also propose a new set of features to model the missingness in clinical data. The pipeline of the proposed method comprises three major components: feature extraction, feature selection, and classification. In total, 407 features are extracted from the clinical data. Then, five different sets of features are selected using a wrapper feature selection algorithm based on XGboost. The selected features are extracted from both valid and missing clinical data. Afterwards, an ensemble model consists of five XGboost models is used for sepsis prediction. The proposed algorithm is ranked officially as third place in the PhysioNet/Computing in Cardiology Challenge 2019 with an overall utility score of 0.339 on the unseen test dataset (our team name: Separatrix).

AB - Sepsis is caused by the dysregulated host response to infection and potentially is the main cause of 6 million death annually. It is a highly dynamic syndrome and therefore the early prediction of sepsis plays a key role in reducing its high associated mortality. However, this is a challenging task because there is no specific and accurate test or scoring system to perform early prediction. In this paper, we present a systematic approach for sepsis prediction. We also propose a new set of features to model the missingness in clinical data. The pipeline of the proposed method comprises three major components: feature extraction, feature selection, and classification. In total, 407 features are extracted from the clinical data. Then, five different sets of features are selected using a wrapper feature selection algorithm based on XGboost. The selected features are extracted from both valid and missing clinical data. Afterwards, an ensemble model consists of five XGboost models is used for sepsis prediction. The proposed algorithm is ranked officially as third place in the PhysioNet/Computing in Cardiology Challenge 2019 with an overall utility score of 0.339 on the unseen test dataset (our team name: Separatrix).

U2 - 10.23919/CinC49843.2019.9005564

DO - 10.23919/CinC49843.2019.9005564

M3 - Conference contribution

T3 - Computing in Cardiology

BT - 2019 Computing in Cardiology, CinC 2019

PB - IEEE Computer Society

ER -