Machine learning-based risk profile classification: A case study for heart valve surgery

Ulrich Bodenhofer; Bettina Haslinger-Eisterer; Alexander Minichmayer; Georg Hermanutz; Jens Meier

Machine learning-based risk profile classification: A case study for heart valve surgery

Ulrich Bodenhofer, Bettina Haslinger-Eisterer, Alexander Minichmayer, Georg Hermanutz, Jens Meier

Research output: Contribution to conference › Paper › peer-review

30 Downloads (Pure)

Abstract

We employ machine learning to predict the 30-days mortality after heart valve surgeries from demographic and preoperative parameters. We achieve AUC values of almost 84%, while the standard EuroSCORE I provides an AUC of only slightly more than 70% for the given cohort. These results indicate (1) that state-of-the-art machine learning is superior to traditional risk models and (2) that calibrating models to specific institutions and surgical procedures allows for more accurate predictions that have the potential to improve medical decision making.

Original language	English (American)
Publication status	Published - Dec 2017
Externally published	Yes
Event	NIPS Workshop on Machine Learning for Health - Long Beach, CA, United States Duration: 8 Dec 2017 → 8 Dec 2017

Workshop

Workshop	NIPS Workshop on Machine Learning for Health
Country/Territory	United States
Period	08.12.2017 → 08.12.2017

Access to Document

BodenhoferHaslinger-EistererMinichmayerHermanutzMeier17Final published version, 148 KB

Cite this

@conference{24bd7d552d6f4886896f36ce55676058,

title = "Machine learning-based risk profile classification: A case study for heart valve surgery",

abstract = "We employ machine learning to predict the 30-days mortality after heart valve surgeries from demographic and preoperative parameters. We achieve AUC values of almost 84%, while the standard EuroSCORE I provides an AUC of only slightly more than 70% for the given cohort. These results indicate (1) that state-of-the-art machine learning is superior to traditional risk models and (2) that calibrating models to specific institutions and surgical procedures allows for more accurate predictions that have the potential to improve medical decision making.",

author = "Ulrich Bodenhofer and Bettina Haslinger-Eisterer and Alexander Minichmayer and Georg Hermanutz and Jens Meier",

year = "2017",

month = dec,

language = "Englisch (Amerika)",

note = "NIPS Workshop on Machine Learning for Health ; Conference date: 08-12-2017 Through 08-12-2017",

}

TY - CONF

T1 - Machine learning-based risk profile classification: A case study for heart valve surgery

AU - Bodenhofer, Ulrich

AU - Haslinger-Eisterer, Bettina

AU - Minichmayer, Alexander

AU - Hermanutz, Georg

AU - Meier, Jens

PY - 2017/12

Y1 - 2017/12

N2 - We employ machine learning to predict the 30-days mortality after heart valve surgeries from demographic and preoperative parameters. We achieve AUC values of almost 84%, while the standard EuroSCORE I provides an AUC of only slightly more than 70% for the given cohort. These results indicate (1) that state-of-the-art machine learning is superior to traditional risk models and (2) that calibrating models to specific institutions and surgical procedures allows for more accurate predictions that have the potential to improve medical decision making.

AB - We employ machine learning to predict the 30-days mortality after heart valve surgeries from demographic and preoperative parameters. We achieve AUC values of almost 84%, while the standard EuroSCORE I provides an AUC of only slightly more than 70% for the given cohort. These results indicate (1) that state-of-the-art machine learning is superior to traditional risk models and (2) that calibrating models to specific institutions and surgical procedures allows for more accurate predictions that have the potential to improve medical decision making.

M3 - Papier

T2 - NIPS Workshop on Machine Learning for Health

Y2 - 8 December 2017 through 8 December 2017

ER -