Optimising coronary imaging decisions with machine learning: an external validation study

Background Exclusion of coronary stenosis in individuals with suggestive symptoms is challenging. Cardiac CT or coronary angiography is often used but is inefficient and costly and involves risks. Sex-stratified algorithms based on electronic health records (EHRs) could be a non-invasive alternative...

Full description

Saved in:
Bibliographic Details
Main Authors: Floor Groepenhoff, Leonard Hofstra, Sophie Heleen Bots, Saskia Haitjema, Imo Hoefer, L. Malin Overmars, Bram van Es, Mark C. H. De Groot, G. Aernout Somsen, I. Igor Tulevski, Hester M. den Ruijter, Wouter W. van Solinge
Format: Article
Language:English
Published: BMJ Publishing Group 2025-05-01
Series:Open Heart
Online Access:https://openheart.bmj.com/content/12/1/e003072.full
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849314290062852096
author Floor Groepenhoff
Leonard Hofstra
Sophie Heleen Bots
Saskia Haitjema
Imo Hoefer
L. Malin Overmars
Bram van Es
Mark C. H. De Groot
G. Aernout Somsen
I. Igor Tulevski
Hester M. den Ruijter
Wouter W. van Solinge
author_facet Floor Groepenhoff
Leonard Hofstra
Sophie Heleen Bots
Saskia Haitjema
Imo Hoefer
L. Malin Overmars
Bram van Es
Mark C. H. De Groot
G. Aernout Somsen
I. Igor Tulevski
Hester M. den Ruijter
Wouter W. van Solinge
author_sort Floor Groepenhoff
collection DOAJ
description Background Exclusion of coronary stenosis in individuals with suggestive symptoms is challenging. Cardiac CT or coronary angiography is often used but is inefficient and costly and involves risks. Sex-stratified algorithms based on electronic health records (EHRs) could be a non-invasive alternative for excluding coronary stenosis, yet their performance may vary by healthcare settings. Thus, external validation is crucial for determining their generalisability. This study aimed to externally validate sex-stratified machine learning algorithms based on EHR data to predict the absence of coronary stenosis, evaluated in diverse clinical settings.Methods Sex-stratified XGBoost algorithms were trained on EHR data from patients who underwent coronary imaging at the University Medical Center Utrecht (n=14 674) and externally tested on EHR data of 13 Cardiology centres in the Netherlands (n=9252). The outcome was defined as the absence of coronary stenosis, identified through text mining of radiology report conclusions, and predictive performance was assessed by negative predictive values (NPVs) and specificities.Results On the training cohort (9298 men (median age 55 years, 73% no coronary stenosis) and 5376 women (median age 59 years, 83% no coronary stenosis)), the algorithms showed NPVs and specificities of 0.95 and 0.14 in men and 0.93 and 0.26 in women, respectively. On the testing cohort (4762 men (median age 60 years, 60% no coronary stenosis) and 4490 women (median age 60 years, 83% no coronary stenosis)), the algorithm showed NPVs and specificities of 0.89 and 0.07 in men and 0.87 and 0.18 in women, respectively.Conclusions This study externally validates sex‐stratified machine learning algorithms using EHR data to non-invasively predict the absence of coronary stenosis, with high NPVs observed across settings. However, given the modest specificity and study limitations, these findings should be considered preliminary, warranting further refinement before clinical adoption.
format Article
id doaj-art-fbaeeed3abd54400a2f8126176c9d5bf
institution Kabale University
issn 2053-3624
language English
publishDate 2025-05-01
publisher BMJ Publishing Group
record_format Article
series Open Heart
spelling doaj-art-fbaeeed3abd54400a2f8126176c9d5bf2025-08-20T03:52:29ZengBMJ Publishing GroupOpen Heart2053-36242025-05-0112110.1136/openhrt-2024-003072Optimising coronary imaging decisions with machine learning: an external validation studyFloor Groepenhoff0Leonard Hofstra1Sophie Heleen Bots2Saskia Haitjema3Imo Hoefer4L. Malin Overmars5Bram van Es6Mark C. H. De Groot7G. Aernout Somsen8I. Igor Tulevski9Hester M. den Ruijter10Wouter W. van Solinge11Central Diagnostic Laboratory, University Medical Center Utrecht, Utrecht, the NetherlandscardiologistLaboratory for Experimental Cardiology, Department of Cardiology, University Medical Center Utrecht, Utrecht, The NetherlandsCentral Diagnostic Laboratory, University Medical Centre Utrecht, Utrecht, The Netherlands3Clinical Diagnostic Laboratory, UMC Utrecht, UtrechtCentral Diagnostic Laboratory, University Medical Centre Utrecht, Utrecht, The NetherlandsCentral Diagnostic Laboratory, University Medical Centre Utrecht, Utrecht, The NetherlandsCentral Diagnostic Laboratory, University Medical Centre Utrecht, Utrecht, The NetherlandsCardiology, Cardiology Centers of the Netherlands, Utrecht, The NetherlandsCardiology, Cardiology Centers of the Netherlands, Amsterdam, The NetherlandsExperimental Cardiology, University Medical Center Utrecht, Utrecht, The NetherlandsCentral Diagnostic Laboratory, University Medical Centre Utrecht, Utrecht, The NetherlandsBackground Exclusion of coronary stenosis in individuals with suggestive symptoms is challenging. Cardiac CT or coronary angiography is often used but is inefficient and costly and involves risks. Sex-stratified algorithms based on electronic health records (EHRs) could be a non-invasive alternative for excluding coronary stenosis, yet their performance may vary by healthcare settings. Thus, external validation is crucial for determining their generalisability. This study aimed to externally validate sex-stratified machine learning algorithms based on EHR data to predict the absence of coronary stenosis, evaluated in diverse clinical settings.Methods Sex-stratified XGBoost algorithms were trained on EHR data from patients who underwent coronary imaging at the University Medical Center Utrecht (n=14 674) and externally tested on EHR data of 13 Cardiology centres in the Netherlands (n=9252). The outcome was defined as the absence of coronary stenosis, identified through text mining of radiology report conclusions, and predictive performance was assessed by negative predictive values (NPVs) and specificities.Results On the training cohort (9298 men (median age 55 years, 73% no coronary stenosis) and 5376 women (median age 59 years, 83% no coronary stenosis)), the algorithms showed NPVs and specificities of 0.95 and 0.14 in men and 0.93 and 0.26 in women, respectively. On the testing cohort (4762 men (median age 60 years, 60% no coronary stenosis) and 4490 women (median age 60 years, 83% no coronary stenosis)), the algorithm showed NPVs and specificities of 0.89 and 0.07 in men and 0.87 and 0.18 in women, respectively.Conclusions This study externally validates sex‐stratified machine learning algorithms using EHR data to non-invasively predict the absence of coronary stenosis, with high NPVs observed across settings. However, given the modest specificity and study limitations, these findings should be considered preliminary, warranting further refinement before clinical adoption.https://openheart.bmj.com/content/12/1/e003072.full
spellingShingle Floor Groepenhoff
Leonard Hofstra
Sophie Heleen Bots
Saskia Haitjema
Imo Hoefer
L. Malin Overmars
Bram van Es
Mark C. H. De Groot
G. Aernout Somsen
I. Igor Tulevski
Hester M. den Ruijter
Wouter W. van Solinge
Optimising coronary imaging decisions with machine learning: an external validation study
Open Heart
title Optimising coronary imaging decisions with machine learning: an external validation study
title_full Optimising coronary imaging decisions with machine learning: an external validation study
title_fullStr Optimising coronary imaging decisions with machine learning: an external validation study
title_full_unstemmed Optimising coronary imaging decisions with machine learning: an external validation study
title_short Optimising coronary imaging decisions with machine learning: an external validation study
title_sort optimising coronary imaging decisions with machine learning an external validation study
url https://openheart.bmj.com/content/12/1/e003072.full
work_keys_str_mv AT floorgroepenhoff optimisingcoronaryimagingdecisionswithmachinelearninganexternalvalidationstudy
AT leonardhofstra optimisingcoronaryimagingdecisionswithmachinelearninganexternalvalidationstudy
AT sophieheleenbots optimisingcoronaryimagingdecisionswithmachinelearninganexternalvalidationstudy
AT saskiahaitjema optimisingcoronaryimagingdecisionswithmachinelearninganexternalvalidationstudy
AT imohoefer optimisingcoronaryimagingdecisionswithmachinelearninganexternalvalidationstudy
AT lmalinovermars optimisingcoronaryimagingdecisionswithmachinelearninganexternalvalidationstudy
AT bramvanes optimisingcoronaryimagingdecisionswithmachinelearninganexternalvalidationstudy
AT markchdegroot optimisingcoronaryimagingdecisionswithmachinelearninganexternalvalidationstudy
AT gaernoutsomsen optimisingcoronaryimagingdecisionswithmachinelearninganexternalvalidationstudy
AT iigortulevski optimisingcoronaryimagingdecisionswithmachinelearninganexternalvalidationstudy
AT hestermdenruijter optimisingcoronaryimagingdecisionswithmachinelearninganexternalvalidationstudy
AT wouterwvansolinge optimisingcoronaryimagingdecisionswithmachinelearninganexternalvalidationstudy