학술논문
Artificial intelligence for detection of ventricular oversensing: Machine learning approaches for noise detection within nonsustained ventricular tachycardia episodes remotely transmitted by pacemakers and implantable cardioverter-defibrillators.
Document Type
Academic Journal
Author
Strik M; Cardio-Thoracic Unit, Bordeaux University Hospital (CHU), Pessac, France; IHU Liryc, Electrophysiology and Heart Modeling Institute, Fondation Bordeaux Université, Université de Bordeaux, Pessac- Bordeaux, France. Electronic address: marcstrik@gmail.com.; Sacristan B; Cardio-Thoracic Unit, Bordeaux University Hospital (CHU), Pessac, France; IHU Liryc, Electrophysiology and Heart Modeling Institute, Fondation Bordeaux Université, Université de Bordeaux, Pessac- Bordeaux, France.; Bordachar P; Cardio-Thoracic Unit, Bordeaux University Hospital (CHU), Pessac, France; IHU Liryc, Electrophysiology and Heart Modeling Institute, Fondation Bordeaux Université, Université de Bordeaux, Pessac- Bordeaux, France.; Duchateau J; Cardio-Thoracic Unit, Bordeaux University Hospital (CHU), Pessac, France; IHU Liryc, Electrophysiology and Heart Modeling Institute, Fondation Bordeaux Université, Université de Bordeaux, Pessac- Bordeaux, France.; Eschalier R; Department of Cardiology, University Hospital Clermont-Ferrand, Clermont-Ferrand, France.; Mondoly P; Department of Cardiology, University Hospital Rangueil, Toulouse, France.; Laborderie J; Department of Cardiology, Bayonne Hospital, Bayonne, France.; Gassa N; IHU Liryc, Electrophysiology and Heart Modeling Institute, Fondation Bordeaux Université, Université de Bordeaux, Pessac- Bordeaux, France.; Zemzemi N; IHU Liryc, Electrophysiology and Heart Modeling Institute, Fondation Bordeaux Université, Université de Bordeaux, Pessac- Bordeaux, France.; Laborde M; IHU Liryc, Electrophysiology and Heart Modeling Institute, Fondation Bordeaux Université, Université de Bordeaux, Pessac- Bordeaux, France.; Garrido J; Universitat Pompeu Fabra, Barcelona, Spain.; Matencio Perabla C; Universitat Pompeu Fabra, Barcelona, Spain.; Jimenez-Perez G; Universitat Pompeu Fabra, Barcelona, Spain.; Camara O; Universitat Pompeu Fabra, Barcelona, Spain.; Haïssaguerre M; Cardio-Thoracic Unit, Bordeaux University Hospital (CHU), Pessac, France; IHU Liryc, Electrophysiology and Heart Modeling Institute, Fondation Bordeaux Université, Université de Bordeaux, Pessac- Bordeaux, France.; Dubois R; IHU Liryc, Electrophysiology and Heart Modeling Institute, Fondation Bordeaux Université, Université de Bordeaux, Pessac- Bordeaux, France.; Ploux S; Cardio-Thoracic Unit, Bordeaux University Hospital (CHU), Pessac, France; IHU Liryc, Electrophysiology and Heart Modeling Institute, Fondation Bordeaux Université, Université de Bordeaux, Pessac- Bordeaux, France.
Source
Publisher: Elsevier Country of Publication: United States NLM ID: 101200317 Publication Model: Print-Electronic Cited Medium: Internet ISSN: 1556-3871 (Electronic) Linking ISSN: 15475271 NLM ISO Abbreviation: Heart Rhythm Subsets: MEDLINE
Subject
Language
English
Abstract
Background: Pacemakers (PMs) and implantable cardioverter-defibrillators (ICDs) increasingly automatically record and remotely transmit nonsustained ventricular tachycardia (NSVT) episodes, which may reveal ventricular oversensing.
Objectives: We aimed to develop and validate a machine learning algorithm that accurately classifies NSVT episodes transmitted by PMs and ICDs in order to lighten health care workload burden and improve patient safety.
Methods: PMs or ICDs (Boston Scientific, St Paul, MN) from 4 French hospitals with ≥1 transmitted NSVT episode were split into 3 subgroups: training set, validation set, and test set. Each NSVT episode was labeled as either physiological or nonphysiological. Four machine learning algorithms-2DTF-CNN, 2D-DenseNet, 2DTF-VGG, and 1D-AgResNet-were developed using training and validation data sets. Accuracies of the classifiers were compared with an analysis of the remote monitoring team of the Bordeaux University Hospital using F2 scores (favoring sensitivity over predictive positive value) using an independent test set.
Results: A total of 807 devices transmitted 10,471 NSVT recordings (82% ICD; 18% PM), of which 87 devices (10.8%) transmitted 544 NSVT recordings with nonphysiological signals. The classification by the remote monitoring team resulted in an F2 score of 0.932 (sensitivity 95%; specificity 99%) The 4 machine learning algorithms showed high and comparable F2 scores (2DTF-CNN: 0.914; 2D-DenseNet: 0.906; 2DTF-VGG: 0.863; 1D-AgResNet: 0.791), and only 1D-AgResNet had significantly different labeling from that of the remote monitoring team.
Conclusion: Machine learning algorithms were accurate in detecting nonphysiological signals within electrograms transmitted by PMs and ICDs. An artificial intelligence approach may render remote monitoring less resourceful and improve patient safety.
(Copyright © 2023 Heart Rhythm Society. Published by Elsevier Inc. All rights reserved.)
Objectives: We aimed to develop and validate a machine learning algorithm that accurately classifies NSVT episodes transmitted by PMs and ICDs in order to lighten health care workload burden and improve patient safety.
Methods: PMs or ICDs (Boston Scientific, St Paul, MN) from 4 French hospitals with ≥1 transmitted NSVT episode were split into 3 subgroups: training set, validation set, and test set. Each NSVT episode was labeled as either physiological or nonphysiological. Four machine learning algorithms-2DTF-CNN, 2D-DenseNet, 2DTF-VGG, and 1D-AgResNet-were developed using training and validation data sets. Accuracies of the classifiers were compared with an analysis of the remote monitoring team of the Bordeaux University Hospital using F2 scores (favoring sensitivity over predictive positive value) using an independent test set.
Results: A total of 807 devices transmitted 10,471 NSVT recordings (82% ICD; 18% PM), of which 87 devices (10.8%) transmitted 544 NSVT recordings with nonphysiological signals. The classification by the remote monitoring team resulted in an F2 score of 0.932 (sensitivity 95%; specificity 99%) The 4 machine learning algorithms showed high and comparable F2 scores (2DTF-CNN: 0.914; 2D-DenseNet: 0.906; 2DTF-VGG: 0.863; 1D-AgResNet: 0.791), and only 1D-AgResNet had significantly different labeling from that of the remote monitoring team.
Conclusion: Machine learning algorithms were accurate in detecting nonphysiological signals within electrograms transmitted by PMs and ICDs. An artificial intelligence approach may render remote monitoring less resourceful and improve patient safety.
(Copyright © 2023 Heart Rhythm Society. Published by Elsevier Inc. All rights reserved.)