Machine Learning to Predict 30 Days and 1-Year Mortality in STEMI and Turndown Patients

Aleeha Iftikhar; RR Bond; Khaled Rjoob; Charles Knoery; Stephen James Leslie; Anne McShane; V. E. McGilligan; Aaron Peace

doi:10.22489/CinC.2020.389

Machine Learning to Predict 30 Days and 1-Year Mortality in STEMI and Turndown Patients

Aleeha Iftikhar, RR Bond, Khaled Rjoob, Charles Knoery, Stephen James Leslie, Anne McShane, V. E. McGilligan, Aaron Peace

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

9 Downloads (Pure)

Abstract

Primary percutaneous coronary intervention (PPCI) is a minimally invasive procedure to unblock the arteries which carry blood to the heart. Referred patients are accepted or turned down for PPCI mainly based on the presence of ST segment elevation on the surface electrocardiogram. We explored the features which predict 30 days and 1-year mortality in accepted and turndown patients and report the performance of machine learning (ML) algorithms. Different ML algorithms, namely multiple logistic regression (MLR), decision tree (DT), and a support vector machine (SVM) were used for the prediction of 30 days and 1-year mortality. Upon significance of various features to predict the 30 days and 1-year mortality, the accuracy, sensitivity, and specificity were compared between algorithms. DT outperformed the other algorithms (SVM and MLR) to predict mortality of patients referred to the PPCI service. Greater sensitivity is achieved in predicting 30 days mortality in the accepted group compared to the turndown group, however, the former model included more features.

Original language	English
Title of host publication	2020 Computing in Cardiology, CinC 2020
Place of Publication	Rimini, Italy
Publisher	IEEE
Pages	1-4
Number of pages	4
Volume	47
Edition	2020
ISBN (Electronic)	978-1-7281-7382-5
ISBN (Print)	978-1-7281-1105-6
DOIs	https://doi.org/10.22489/CinC.2020.389
Publication status	Published (in print/issue) - 10 Feb 2021
Event	Computing in Cardiology 2020 - Palacongressi, Rimini, Italy Duration: 13 Sept 2020 → 16 Sept 2020

Publication series

Name	2020 COMPUTING IN CARDIOLOGY
ISSN (Print)	2325-8861

Conference

Conference	Computing in Cardiology 2020
Abbreviated title	CinC20
Country/Territory	Italy
City	Rimini
Period	13/09/20 → 16/09/20

Bibliographical note

Funding Information:
The author(s) disclosed receipt of the following financial support for the research, authorship and/or publication of this article: This research is supported by the European Union’s INTERREG VA Programme, managed by the Special EU Programmes Body (SEUPB).

Publisher Copyright:
© 2020 Creative Commons; the authors hold their copyright.

Keywords

Machine Learning
PPCI Referral
cathlab
cardiology
angioplasty

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.22489/CinC.2020.389

CINC_Machine learning to predict 30 days and 1-year mortality in STEMI and turndown patients submittedAuthor Accepted Manuscript, 72 KB

Cite this

@inproceedings{9a12ffb846d24423bfb44e3b03a94cfb,

title = "Machine Learning to Predict 30 Days and 1-Year Mortality in STEMI and Turndown Patients",

abstract = "Primary percutaneous coronary intervention (PPCI) is a minimally invasive procedure to unblock the arteries which carry blood to the heart. Referred patients are accepted or turned down for PPCI mainly based on the presence of ST segment elevation on the surface electrocardiogram. We explored the features which predict 30 days and 1-year mortality in accepted and turndown patients and report the performance of machine learning (ML) algorithms. Different ML algorithms, namely multiple logistic regression (MLR), decision tree (DT), and a support vector machine (SVM) were used for the prediction of 30 days and 1-year mortality. Upon significance of various features to predict the 30 days and 1-year mortality, the accuracy, sensitivity, and specificity were compared between algorithms. DT outperformed the other algorithms (SVM and MLR) to predict mortality of patients referred to the PPCI service. Greater sensitivity is achieved in predicting 30 days mortality in the accepted group compared to the turndown group, however, the former model included more features.",

keywords = "Machine Learning, PPCI Referral, cathlab, cardiology, angioplasty",

author = "Aleeha Iftikhar and RR Bond and Khaled Rjoob and Charles Knoery and Leslie, {Stephen James} and Anne McShane and McGilligan, {V. E.} and Aaron Peace",

note = "Funding Information: The author(s) disclosed receipt of the following financial support for the research, authorship and/or publication of this article: This research is supported by the European Union{\textquoteright}s INTERREG VA Programme, managed by the Special EU Programmes Body (SEUPB). Publisher Copyright: {\textcopyright} 2020 Creative Commons; the authors hold their copyright.; Computing in Cardiology 2020, CinC20 ; Conference date: 13-09-2020 Through 16-09-2020",

year = "2021",

month = feb,

day = "10",

doi = "10.22489/CinC.2020.389",

language = "English",

isbn = "978-1-7281-1105-6",

volume = "47",

series = "2020 COMPUTING IN CARDIOLOGY",

publisher = "IEEE",

pages = "1--4",

booktitle = "2020 Computing in Cardiology, CinC 2020",

address = "United States",

edition = "2020",

}

Iftikhar, A, Bond, RR, Rjoob, K, Knoery, C, Leslie, SJ, McShane, A, McGilligan, VE & Peace, A 2021, Machine Learning to Predict 30 Days and 1-Year Mortality in STEMI and Turndown Patients. in 2020 Computing in Cardiology, CinC 2020. 2020 edn, vol. 47, 9344181, 2020 COMPUTING IN CARDIOLOGY, IEEE, Rimini, Italy, pp. 1-4, Computing in Cardiology 2020, Rimini, Italy, 13/09/20. https://doi.org/10.22489/CinC.2020.389

Machine Learning to Predict 30 Days and 1-Year Mortality in STEMI and Turndown Patients. / Iftikhar, Aleeha; Bond, RR; Rjoob, Khaled et al.
2020 Computing in Cardiology, CinC 2020. Vol. 47 2020. ed. Rimini, Italy: IEEE, 2021. p. 1-4 9344181 (2020 COMPUTING IN CARDIOLOGY).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Machine Learning to Predict 30 Days and 1-Year Mortality in STEMI and Turndown Patients

AU - Iftikhar, Aleeha

AU - Bond, RR

AU - Rjoob, Khaled

AU - Knoery, Charles

AU - Leslie, Stephen James

AU - McShane, Anne

AU - McGilligan, V. E.

AU - Peace, Aaron

N1 - Funding Information: The author(s) disclosed receipt of the following financial support for the research, authorship and/or publication of this article: This research is supported by the European Union’s INTERREG VA Programme, managed by the Special EU Programmes Body (SEUPB). Publisher Copyright: © 2020 Creative Commons; the authors hold their copyright.

PY - 2021/2/10

Y1 - 2021/2/10

N2 - Primary percutaneous coronary intervention (PPCI) is a minimally invasive procedure to unblock the arteries which carry blood to the heart. Referred patients are accepted or turned down for PPCI mainly based on the presence of ST segment elevation on the surface electrocardiogram. We explored the features which predict 30 days and 1-year mortality in accepted and turndown patients and report the performance of machine learning (ML) algorithms. Different ML algorithms, namely multiple logistic regression (MLR), decision tree (DT), and a support vector machine (SVM) were used for the prediction of 30 days and 1-year mortality. Upon significance of various features to predict the 30 days and 1-year mortality, the accuracy, sensitivity, and specificity were compared between algorithms. DT outperformed the other algorithms (SVM and MLR) to predict mortality of patients referred to the PPCI service. Greater sensitivity is achieved in predicting 30 days mortality in the accepted group compared to the turndown group, however, the former model included more features.

AB - Primary percutaneous coronary intervention (PPCI) is a minimally invasive procedure to unblock the arteries which carry blood to the heart. Referred patients are accepted or turned down for PPCI mainly based on the presence of ST segment elevation on the surface electrocardiogram. We explored the features which predict 30 days and 1-year mortality in accepted and turndown patients and report the performance of machine learning (ML) algorithms. Different ML algorithms, namely multiple logistic regression (MLR), decision tree (DT), and a support vector machine (SVM) were used for the prediction of 30 days and 1-year mortality. Upon significance of various features to predict the 30 days and 1-year mortality, the accuracy, sensitivity, and specificity were compared between algorithms. DT outperformed the other algorithms (SVM and MLR) to predict mortality of patients referred to the PPCI service. Greater sensitivity is achieved in predicting 30 days mortality in the accepted group compared to the turndown group, however, the former model included more features.

KW - Machine Learning

KW - PPCI Referral

KW - cathlab

KW - cardiology

KW - angioplasty

UR - https://ieeexplore.ieee.org/document/9344181

UR - http://www.scopus.com/inward/record.url?scp=85100917690&partnerID=8YFLogxK

U2 - 10.22489/CinC.2020.389

DO - 10.22489/CinC.2020.389

M3 - Conference contribution

SN - 978-1-7281-1105-6

VL - 47

T3 - 2020 COMPUTING IN CARDIOLOGY

SP - 1

EP - 4

BT - 2020 Computing in Cardiology, CinC 2020

PB - IEEE

CY - Rimini, Italy

T2 - Computing in Cardiology 2020

Y2 - 13 September 2020 through 16 September 2020

ER -

Machine Learning to Predict 30 Days and 1-Year Mortality in STEMI and Turndown Patients

Abstract

Publication series

Conference

Bibliographical note

Keywords

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this