Cover Image

Predicting 30-day mortality after an acute coronary syndrome (ACS) using machine learning methods for feature selection, classification and visualization

Hybrid combinations of feature selection, classification and visualisation using machine learning (ML) methods have the potential for enhanced understanding and 30-day mortality prediction of patients with cardiovascular disease using population-specific data. Identifying a feature selection met...

Full description

Saved in:
Bibliographic Details
Main Authors: Nanyonga Aziida,, Sorayya Malek,, Firdaus Aziz,, Khairul Shafiq Ibrahim,, Sazzli Kasim,
Format: Article
Language:English
Published: Penerbit Universiti Kebangsaan Malaysia 2021
Online Access:http://journalarticle.ukm.my/16915/1/17.pdf
http://journalarticle.ukm.my/16915/
https://www.ukm.my/jsm/malay_journals/jilid50bil3_2021/KandunganJilid50Bil3_2021.html
Tags: Add Tag
No Tags, Be the first to tag this record!
id my-ukm.journal.16915
record_format eprints
spelling my-ukm.journal.169152021-06-28T15:32:54Z http://journalarticle.ukm.my/16915/ Predicting 30-day mortality after an acute coronary syndrome (ACS) using machine learning methods for feature selection, classification and visualization Nanyonga Aziida, Sorayya Malek, Firdaus Aziz, Khairul Shafiq Ibrahim, Sazzli Kasim, Hybrid combinations of feature selection, classification and visualisation using machine learning (ML) methods have the potential for enhanced understanding and 30-day mortality prediction of patients with cardiovascular disease using population-specific data. Identifying a feature selection method with a classifier algorithm that produces high performance in mortality studies is essential and has not been reported before. Feature selection methods such as Boruta, Random Forest (RF), Elastic Net (EN), Recursive Feature Elimination (RFE), learning vector quantization (LVQ), Genetic Algorithm (GA), Cluster Dendrogram (CD), Support Vector Machine (SVM) and Logistic Regression (LR) were combined with RF, SVM, LR, and EN classifiers for 30-day mortality prediction. ML models were constructed using 302 patients and 54 input variables from the Malaysian National Cardiovascular Disease Database. Validation of the best ML model was performed against Thrombolysis in Myocardial Infarction (TIMI) using an additional dataset of 102 patients. The Self-Organising Feature Map (SOM) was used to visualise mortality-related factors post-ACS. The performance of ML models using the area under the curve (AUC) ranged from 0.48 to 0.80. The best-performing model (AUC = 0.80) was a hybrid combination of the RF variable importance method, the sequential backward selection and the RF classifier using five predictors (age, triglyceride, creatinine, troponin, and total cholesterol). Comparison with TIMI using an additional dataset resulted in the best ML model outperforming the TIMI score (AUC = 0.75 vs. AUC = 0.60). The findings of this study will provide a basis for developing an online ML-based population-specific risk scoring calculator. Penerbit Universiti Kebangsaan Malaysia 2021-03 Article PeerReviewed application/pdf en http://journalarticle.ukm.my/16915/1/17.pdf Nanyonga Aziida, and Sorayya Malek, and Firdaus Aziz, and Khairul Shafiq Ibrahim, and Sazzli Kasim, (2021) Predicting 30-day mortality after an acute coronary syndrome (ACS) using machine learning methods for feature selection, classification and visualization. Sains Malaysiana, 50 (3). pp. 753-768. ISSN 0126-6039 https://www.ukm.my/jsm/malay_journals/jilid50bil3_2021/KandunganJilid50Bil3_2021.html
institution Universiti Kebangsaan Malaysia
building Tun Sri Lanang Library
collection Institutional Repository
continent Asia
country Malaysia
content_provider Universiti Kebangsaan Malaysia
content_source UKM Journal Article Repository
url_provider http://journalarticle.ukm.my/
language English
description Hybrid combinations of feature selection, classification and visualisation using machine learning (ML) methods have the potential for enhanced understanding and 30-day mortality prediction of patients with cardiovascular disease using population-specific data. Identifying a feature selection method with a classifier algorithm that produces high performance in mortality studies is essential and has not been reported before. Feature selection methods such as Boruta, Random Forest (RF), Elastic Net (EN), Recursive Feature Elimination (RFE), learning vector quantization (LVQ), Genetic Algorithm (GA), Cluster Dendrogram (CD), Support Vector Machine (SVM) and Logistic Regression (LR) were combined with RF, SVM, LR, and EN classifiers for 30-day mortality prediction. ML models were constructed using 302 patients and 54 input variables from the Malaysian National Cardiovascular Disease Database. Validation of the best ML model was performed against Thrombolysis in Myocardial Infarction (TIMI) using an additional dataset of 102 patients. The Self-Organising Feature Map (SOM) was used to visualise mortality-related factors post-ACS. The performance of ML models using the area under the curve (AUC) ranged from 0.48 to 0.80. The best-performing model (AUC = 0.80) was a hybrid combination of the RF variable importance method, the sequential backward selection and the RF classifier using five predictors (age, triglyceride, creatinine, troponin, and total cholesterol). Comparison with TIMI using an additional dataset resulted in the best ML model outperforming the TIMI score (AUC = 0.75 vs. AUC = 0.60). The findings of this study will provide a basis for developing an online ML-based population-specific risk scoring calculator.
format Article
author Nanyonga Aziida,
Sorayya Malek,
Firdaus Aziz,
Khairul Shafiq Ibrahim,
Sazzli Kasim,
spellingShingle Nanyonga Aziida,
Sorayya Malek,
Firdaus Aziz,
Khairul Shafiq Ibrahim,
Sazzli Kasim,
Predicting 30-day mortality after an acute coronary syndrome (ACS) using machine learning methods for feature selection, classification and visualization
author_facet Nanyonga Aziida,
Sorayya Malek,
Firdaus Aziz,
Khairul Shafiq Ibrahim,
Sazzli Kasim,
author_sort Nanyonga Aziida,
title Predicting 30-day mortality after an acute coronary syndrome (ACS) using machine learning methods for feature selection, classification and visualization
title_short Predicting 30-day mortality after an acute coronary syndrome (ACS) using machine learning methods for feature selection, classification and visualization
title_full Predicting 30-day mortality after an acute coronary syndrome (ACS) using machine learning methods for feature selection, classification and visualization
title_fullStr Predicting 30-day mortality after an acute coronary syndrome (ACS) using machine learning methods for feature selection, classification and visualization
title_full_unstemmed Predicting 30-day mortality after an acute coronary syndrome (ACS) using machine learning methods for feature selection, classification and visualization
title_sort predicting 30-day mortality after an acute coronary syndrome (acs) using machine learning methods for feature selection, classification and visualization
publisher Penerbit Universiti Kebangsaan Malaysia
publishDate 2021
url http://journalarticle.ukm.my/16915/1/17.pdf
http://journalarticle.ukm.my/16915/
https://www.ukm.my/jsm/malay_journals/jilid50bil3_2021/KandunganJilid50Bil3_2021.html
_version_ 1703961608342994944
score 13.211869

Similar Items