Prior electrocardiograms not useful for machine learning predictions of major adverse cardiac events in emergency department chest pain patients
(2024) In Journal of Electrocardiology 82. p.42-51- Abstract
- At the emergency department (ED), it is important to quickly and accurately determine which patients are likely to have a major adverse cardiac event (MACE). Machine learning (ML) models can be used to aid physicians in detecting MACE, and improving the performance of such models is an active area of research. In this study, we sought to determine if ML models can be improved by including a prior electrocardiogram (ECG) from each patient. To that end, we trained several models to predict MACE within 30 days, both with and without prior ECGs, using data collected from 19,499 consecutive patients with chest pain, from five EDs in southern Sweden, between the years 2017 and 2018. Our results indicate no improvement in AUC from prior ECGs.... (More)
- At the emergency department (ED), it is important to quickly and accurately determine which patients are likely to have a major adverse cardiac event (MACE). Machine learning (ML) models can be used to aid physicians in detecting MACE, and improving the performance of such models is an active area of research. In this study, we sought to determine if ML models can be improved by including a prior electrocardiogram (ECG) from each patient. To that end, we trained several models to predict MACE within 30 days, both with and without prior ECGs, using data collected from 19,499 consecutive patients with chest pain, from five EDs in southern Sweden, between the years 2017 and 2018. Our results indicate no improvement in AUC from prior ECGs. This was consistent across models, both with and without additional clinical input variables, for different patient subgroups, and for different subsets of the outcome. While contradicting current best practices for manual ECG analysis, the results are positive in the sense that ML models with fewer inputs are more easily and widely applicable in practice. (Less)
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/7d494e5b-91a0-4e58-a76c-952766793e4a
- author
- Nyström, Axel LU ; Olsson de Capretz, Pontus LU ; Björkelund, Anders LU ; Lundager Forberg, Jakob LU ; Ohlsson, Mattias LU ; Björk, Jonas LU and Ekelund, Ulf LU
- organization
-
- EPI@LUND (research group)
- Division of Occupational and Environmental Medicine, Lund University
- Computational Science for Health and Environment (research group)
- Emergency medicine (research group)
- Centre for Environmental and Climate Science (CEC)
- LU Profile Area: Natural and Artificial Cognition
- eSSENCE: The e-Science Collaboration
- Artificial Intelligence in CardioThoracic Sciences (AICTS) (research group)
- Surgery and public health (research group)
- EpiHealth: Epidemiology for Health
- publishing date
- 2024
- type
- Contribution to journal
- publication status
- published
- subject
- in
- Journal of Electrocardiology
- volume
- 82
- pages
- 42 - 51
- publisher
- Elsevier
- external identifiers
-
- pmid:38006763
- scopus:85182224199
- ISSN
- 1532-8430
- DOI
- 10.1016/j.jelectrocard.2023.11.002
- project
- AIR Lund - Artificially Intelligent use of Registers
- language
- English
- LU publication?
- yes
- id
- 7d494e5b-91a0-4e58-a76c-952766793e4a
- date added to LUP
- 2024-01-02 12:25:02
- date last changed
- 2024-04-17 13:23:33
@article{7d494e5b-91a0-4e58-a76c-952766793e4a, abstract = {{At the emergency department (ED), it is important to quickly and accurately determine which patients are likely to have a major adverse cardiac event (MACE). Machine learning (ML) models can be used to aid physicians in detecting MACE, and improving the performance of such models is an active area of research. In this study, we sought to determine if ML models can be improved by including a prior electrocardiogram (ECG) from each patient. To that end, we trained several models to predict MACE within 30 days, both with and without prior ECGs, using data collected from 19,499 consecutive patients with chest pain, from five EDs in southern Sweden, between the years 2017 and 2018. Our results indicate no improvement in AUC from prior ECGs. This was consistent across models, both with and without additional clinical input variables, for different patient subgroups, and for different subsets of the outcome. While contradicting current best practices for manual ECG analysis, the results are positive in the sense that ML models with fewer inputs are more easily and widely applicable in practice.}}, author = {{Nyström, Axel and Olsson de Capretz, Pontus and Björkelund, Anders and Lundager Forberg, Jakob and Ohlsson, Mattias and Björk, Jonas and Ekelund, Ulf}}, issn = {{1532-8430}}, language = {{eng}}, pages = {{42--51}}, publisher = {{Elsevier}}, series = {{Journal of Electrocardiology}}, title = {{Prior electrocardiograms not useful for machine learning predictions of major adverse cardiac events in emergency department chest pain patients}}, url = {{http://dx.doi.org/10.1016/j.jelectrocard.2023.11.002}}, doi = {{10.1016/j.jelectrocard.2023.11.002}}, volume = {{82}}, year = {{2024}}, }