Development and validation of machine learning-derived frailty index in predicting outcomes of patients undergoing percutaneous coronary intervention

Introduction: Frailty is associated with increased mortality in patients with percutaneous coronary intervention (PCI). Existing operationalized frailty measurement tools are limited and require resource intensive process. We developed and validated a tool to identify and stratify frailty using coll...

Full description

Saved in:
Bibliographic Details
Main Authors: John T.Y. Soong, L.F. Tan, Rodney Y.H. Soh, W.B. He, Andie H. Djohan, H.W. Sim, T.C. Yeo, H.C. Tan, Mark Y.Y. Chan, C.H. Sia, M.L. Feng
Format: Article
Language:English
Published: Elsevier 2024-12-01
Series:International Journal of Cardiology: Heart & Vasculature
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2352906724001775
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850065603263463424
author John T.Y. Soong
L.F. Tan
Rodney Y.H. Soh
W.B. He
Andie H. Djohan
H.W. Sim
T.C. Yeo
H.C. Tan
Mark Y.Y. Chan
C.H. Sia
M.L. Feng
author_facet John T.Y. Soong
L.F. Tan
Rodney Y.H. Soh
W.B. He
Andie H. Djohan
H.W. Sim
T.C. Yeo
H.C. Tan
Mark Y.Y. Chan
C.H. Sia
M.L. Feng
author_sort John T.Y. Soong
collection DOAJ
description Introduction: Frailty is associated with increased mortality in patients with percutaneous coronary intervention (PCI). Existing operationalized frailty measurement tools are limited and require resource intensive process. We developed and validated a tool to identify and stratify frailty using collected data for patients who underwent PCI and explored its predictive power to predict adverse clinical outcomes post PCI. Methods: Between 2014 and 2015, 1,732 patients who underwent semi-urgent or elective PCI in a tertiary centre were included. Variables including demographics, co-morbidities, investigations and clinical outcomes to 33 ± 37 months were analysed. Logistic regression model and Extreme Gradient Boosting (XGBoost) machine learning model were constructed to identify predictors of adverse clinical outcomes post PCI. The final models’ predicted probabilities were assessed with area under receiver operating characteristic curve (AUC). Results: With model analysis, frailty index (FI), age and gender were the 3 most important features for adverse clinical outcomes prediction, with FI contributing the most. After adjustment, the odds of FI to predict cardiac death and in-hospital death post PCI remained significant [1.94 (95 %CI1.79–2.10); p < 0.001, 2.04(95 %CI 1.87–2.23); p < 0.001 respectively]. The XGBoost machine learning models improved predictive power for cardiac death [AUC 0.83(95 %CI 0.80–0.86)] and in hospital death [AUC 0.83(95 %CI 0.80–0.86)] post PCI compared to logistic regression models. Conclusion: The resultant model developed using novel machine learning methodologies had good predictive power for significant clinical outcomes post PCI with potential to be automated within hospital information systems.
format Article
id doaj-art-7a9057b60fc74897b91d92c93bcc27a9
institution DOAJ
issn 2352-9067
language English
publishDate 2024-12-01
publisher Elsevier
record_format Article
series International Journal of Cardiology: Heart & Vasculature
spelling doaj-art-7a9057b60fc74897b91d92c93bcc27a92025-08-20T02:48:57ZengElsevierInternational Journal of Cardiology: Heart & Vasculature2352-90672024-12-015510151110.1016/j.ijcha.2024.101511Development and validation of machine learning-derived frailty index in predicting outcomes of patients undergoing percutaneous coronary interventionJohn T.Y. Soong0L.F. Tan1Rodney Y.H. Soh2W.B. He3Andie H. Djohan4H.W. Sim5T.C. Yeo6H.C. Tan7Mark Y.Y. Chan8C.H. Sia9M.L. Feng10Yong Loo Lin School of Medicine, National University Singapore, Department of Medicine, National University Hospital, SingaporeDepartment of Medicine, National University Hospital, Singapore, Alexandra Hospital, SingaporeNational University Heart Centre, National University Hospital, Singapore; Corresponding authors.Institute of Hospital Management, West China Hospital of Sichuan University, Chengdu, Sichuan, China; Institute of Data Science, National University of Singapore, Singapore; Corresponding authors.National University Heart Centre, National University Hospital, SingaporeNational University Heart Centre, National University Hospital, SingaporeYong Loo Lin School of Medicine, National University Singapore, National University Heart Centre, National University Hospital, SingaporeYong Loo Lin School of Medicine, National University Singapore, National University Heart Centre, National University Hospital, SingaporeYong Loo Lin School of Medicine, National University Singapore, National University Heart Centre, National University Hospital, SingaporeYong Loo Lin School of Medicine, National University Singapore, National University Heart Centre, National University Hospital, SingaporeSaw Swee Hock School of Public Health, National University Health System, Institute of Data Science, National University of Singapore, SingaporeIntroduction: Frailty is associated with increased mortality in patients with percutaneous coronary intervention (PCI). Existing operationalized frailty measurement tools are limited and require resource intensive process. We developed and validated a tool to identify and stratify frailty using collected data for patients who underwent PCI and explored its predictive power to predict adverse clinical outcomes post PCI. Methods: Between 2014 and 2015, 1,732 patients who underwent semi-urgent or elective PCI in a tertiary centre were included. Variables including demographics, co-morbidities, investigations and clinical outcomes to 33 ± 37 months were analysed. Logistic regression model and Extreme Gradient Boosting (XGBoost) machine learning model were constructed to identify predictors of adverse clinical outcomes post PCI. The final models’ predicted probabilities were assessed with area under receiver operating characteristic curve (AUC). Results: With model analysis, frailty index (FI), age and gender were the 3 most important features for adverse clinical outcomes prediction, with FI contributing the most. After adjustment, the odds of FI to predict cardiac death and in-hospital death post PCI remained significant [1.94 (95 %CI1.79–2.10); p < 0.001, 2.04(95 %CI 1.87–2.23); p < 0.001 respectively]. The XGBoost machine learning models improved predictive power for cardiac death [AUC 0.83(95 %CI 0.80–0.86)] and in hospital death [AUC 0.83(95 %CI 0.80–0.86)] post PCI compared to logistic regression models. Conclusion: The resultant model developed using novel machine learning methodologies had good predictive power for significant clinical outcomes post PCI with potential to be automated within hospital information systems.http://www.sciencedirect.com/science/article/pii/S2352906724001775Novel machine learningFrailty indexPercutaneous coronary interventionAsian
spellingShingle John T.Y. Soong
L.F. Tan
Rodney Y.H. Soh
W.B. He
Andie H. Djohan
H.W. Sim
T.C. Yeo
H.C. Tan
Mark Y.Y. Chan
C.H. Sia
M.L. Feng
Development and validation of machine learning-derived frailty index in predicting outcomes of patients undergoing percutaneous coronary intervention
International Journal of Cardiology: Heart & Vasculature
Novel machine learning
Frailty index
Percutaneous coronary intervention
Asian
title Development and validation of machine learning-derived frailty index in predicting outcomes of patients undergoing percutaneous coronary intervention
title_full Development and validation of machine learning-derived frailty index in predicting outcomes of patients undergoing percutaneous coronary intervention
title_fullStr Development and validation of machine learning-derived frailty index in predicting outcomes of patients undergoing percutaneous coronary intervention
title_full_unstemmed Development and validation of machine learning-derived frailty index in predicting outcomes of patients undergoing percutaneous coronary intervention
title_short Development and validation of machine learning-derived frailty index in predicting outcomes of patients undergoing percutaneous coronary intervention
title_sort development and validation of machine learning derived frailty index in predicting outcomes of patients undergoing percutaneous coronary intervention
topic Novel machine learning
Frailty index
Percutaneous coronary intervention
Asian
url http://www.sciencedirect.com/science/article/pii/S2352906724001775
work_keys_str_mv AT johntysoong developmentandvalidationofmachinelearningderivedfrailtyindexinpredictingoutcomesofpatientsundergoingpercutaneouscoronaryintervention
AT lftan developmentandvalidationofmachinelearningderivedfrailtyindexinpredictingoutcomesofpatientsundergoingpercutaneouscoronaryintervention
AT rodneyyhsoh developmentandvalidationofmachinelearningderivedfrailtyindexinpredictingoutcomesofpatientsundergoingpercutaneouscoronaryintervention
AT wbhe developmentandvalidationofmachinelearningderivedfrailtyindexinpredictingoutcomesofpatientsundergoingpercutaneouscoronaryintervention
AT andiehdjohan developmentandvalidationofmachinelearningderivedfrailtyindexinpredictingoutcomesofpatientsundergoingpercutaneouscoronaryintervention
AT hwsim developmentandvalidationofmachinelearningderivedfrailtyindexinpredictingoutcomesofpatientsundergoingpercutaneouscoronaryintervention
AT tcyeo developmentandvalidationofmachinelearningderivedfrailtyindexinpredictingoutcomesofpatientsundergoingpercutaneouscoronaryintervention
AT hctan developmentandvalidationofmachinelearningderivedfrailtyindexinpredictingoutcomesofpatientsundergoingpercutaneouscoronaryintervention
AT markyychan developmentandvalidationofmachinelearningderivedfrailtyindexinpredictingoutcomesofpatientsundergoingpercutaneouscoronaryintervention
AT chsia developmentandvalidationofmachinelearningderivedfrailtyindexinpredictingoutcomesofpatientsundergoingpercutaneouscoronaryintervention
AT mlfeng developmentandvalidationofmachinelearningderivedfrailtyindexinpredictingoutcomesofpatientsundergoingpercutaneouscoronaryintervention