Machine learning and feature engineering for predicting pulse presence during chest compressions

Current resuscitation protocols require pausing chest compressions during cardiopulmonary resuscitation (CPR) to check for a pulse. However, pausing CPR when a patient is pulseless can worsen patient outcomes. Our objective was to design and evaluate an ECG-based algorithm that predicts pulse presen...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Diya Sashidhar, Heemun Kwok, Jason Coult, Jennifer Blackwood, Peter J. Kudenchuk, Shiv Bhandari, Thomas D. Rea, J. Nathan Kutz
Formato:	article
Lenguaje:	EN
Publicado:	The Royal Society 2021
Materias:	wavelets machine learning resuscitation Science Q
Acceso en línea:	https://doaj.org/article/60d420c517094551a0cee642b8ab944b
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:60d420c517094551a0cee642b8ab944b
record_format	dspace
spelling	oai:doaj.org-article:60d420c517094551a0cee642b8ab944b2021-11-10T08:06:33ZMachine learning and feature engineering for predicting pulse presence during chest compressions10.1098/rsos.2105662054-5703https://doaj.org/article/60d420c517094551a0cee642b8ab944b2021-11-01T00:00:00Zhttps://royalsocietypublishing.org/doi/10.1098/rsos.210566https://doaj.org/toc/2054-5703Current resuscitation protocols require pausing chest compressions during cardiopulmonary resuscitation (CPR) to check for a pulse. However, pausing CPR when a patient is pulseless can worsen patient outcomes. Our objective was to design and evaluate an ECG-based algorithm that predicts pulse presence with or without CPR. We evaluated 383 patients being treated for out-of-hospital cardiac arrest with real-time ECG, impedance and audio recordings. Paired ECG segments having an organized rhythm immediately preceding a pulse check (during CPR) and during the pulse check (without CPR) were extracted. Patients were randomly divided into 60% training and 40% test groups. From training data, we developed an algorithm to predict the clinical pulse presence based on the wavelet transform of the bandpass-filtered ECG. Principal component analysis was used to reduce dimensionality, and we then trained a linear discriminant model using three principal component modes as input features. Overall, 38% (351/912) of checks had a spontaneous pulse. AUCs for predicting pulse presence with and without CPR on test data were 0.84 (95% CI (0.80, 0.88)) and 0.89 (95% CI (0.86, 0.92)), respectively. This ECG-based algorithm demonstrates potential to improve resuscitation by predicting the presence of a spontaneous pulse without pausing CPR with moderate accuracy.Diya SashidharHeemun KwokJason CoultJennifer BlackwoodPeter J. KudenchukShiv BhandariThomas D. ReaJ. Nathan KutzThe Royal Societyarticlewaveletsmachine learningresuscitationScienceQENRoyal Society Open Science, Vol 8, Iss 11 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	wavelets machine learning resuscitation Science Q
spellingShingle	wavelets machine learning resuscitation Science Q Diya Sashidhar Heemun Kwok Jason Coult Jennifer Blackwood Peter J. Kudenchuk Shiv Bhandari Thomas D. Rea J. Nathan Kutz Machine learning and feature engineering for predicting pulse presence during chest compressions
description	Current resuscitation protocols require pausing chest compressions during cardiopulmonary resuscitation (CPR) to check for a pulse. However, pausing CPR when a patient is pulseless can worsen patient outcomes. Our objective was to design and evaluate an ECG-based algorithm that predicts pulse presence with or without CPR. We evaluated 383 patients being treated for out-of-hospital cardiac arrest with real-time ECG, impedance and audio recordings. Paired ECG segments having an organized rhythm immediately preceding a pulse check (during CPR) and during the pulse check (without CPR) were extracted. Patients were randomly divided into 60% training and 40% test groups. From training data, we developed an algorithm to predict the clinical pulse presence based on the wavelet transform of the bandpass-filtered ECG. Principal component analysis was used to reduce dimensionality, and we then trained a linear discriminant model using three principal component modes as input features. Overall, 38% (351/912) of checks had a spontaneous pulse. AUCs for predicting pulse presence with and without CPR on test data were 0.84 (95% CI (0.80, 0.88)) and 0.89 (95% CI (0.86, 0.92)), respectively. This ECG-based algorithm demonstrates potential to improve resuscitation by predicting the presence of a spontaneous pulse without pausing CPR with moderate accuracy.
format	article
author	Diya Sashidhar Heemun Kwok Jason Coult Jennifer Blackwood Peter J. Kudenchuk Shiv Bhandari Thomas D. Rea J. Nathan Kutz
author_facet	Diya Sashidhar Heemun Kwok Jason Coult Jennifer Blackwood Peter J. Kudenchuk Shiv Bhandari Thomas D. Rea J. Nathan Kutz
author_sort	Diya Sashidhar
title	Machine learning and feature engineering for predicting pulse presence during chest compressions
title_short	Machine learning and feature engineering for predicting pulse presence during chest compressions
title_full	Machine learning and feature engineering for predicting pulse presence during chest compressions
title_fullStr	Machine learning and feature engineering for predicting pulse presence during chest compressions
title_full_unstemmed	Machine learning and feature engineering for predicting pulse presence during chest compressions
title_sort	machine learning and feature engineering for predicting pulse presence during chest compressions
publisher	The Royal Society
publishDate	2021
url	https://doaj.org/article/60d420c517094551a0cee642b8ab944b
work_keys_str_mv	AT diyasashidhar machinelearningandfeatureengineeringforpredictingpulsepresenceduringchestcompressions AT heemunkwok machinelearningandfeatureengineeringforpredictingpulsepresenceduringchestcompressions AT jasoncoult machinelearningandfeatureengineeringforpredictingpulsepresenceduringchestcompressions AT jenniferblackwood machinelearningandfeatureengineeringforpredictingpulsepresenceduringchestcompressions AT peterjkudenchuk machinelearningandfeatureengineeringforpredictingpulsepresenceduringchestcompressions AT shivbhandari machinelearningandfeatureengineeringforpredictingpulsepresenceduringchestcompressions AT thomasdrea machinelearningandfeatureengineeringforpredictingpulsepresenceduringchestcompressions AT jnathankutz machinelearningandfeatureengineeringforpredictingpulsepresenceduringchestcompressions
_version_	1718440377834274816

Machine learning and feature engineering for predicting pulse presence during chest compressions

Ejemplares similares