Convolution Neural Network Algorithm for Shockable Arrhythmia Classification Within a Digitally Connected Automated External Defibrillator
Background Diagnosis of shockable rhythms leading to defibrillation remains integral to improving out‐of‐hospital cardiac arrest outcomes. New machine learning techniques have emerged to diagnose arrhythmias on ECGs. In out‐of‐hospital cardiac arrest, an algorithm within an automated external defibr...
| Main Authors: | , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2023-04-01
|
| Series: | Journal of the American Heart Association: Cardiovascular and Cerebrovascular Disease |
| Subjects: | |
| Online Access: | https://www.ahajournals.org/doi/10.1161/JAHA.122.026974 |
| _version_ | 1797662589450190848 |
|---|---|
| author | Christine P. Shen Benjamin C. Freed David P. Walter James C. Perry Amr F. Barakat Ahmad Ramy A. Elashery Kevin S. Shah Shelby Kutty Michael McGillion Fu Siong Ng Rola Khedraki Keshav R. Nayak John D. Rogers Sanjeev P. Bhavnani |
| author_facet | Christine P. Shen Benjamin C. Freed David P. Walter James C. Perry Amr F. Barakat Ahmad Ramy A. Elashery Kevin S. Shah Shelby Kutty Michael McGillion Fu Siong Ng Rola Khedraki Keshav R. Nayak John D. Rogers Sanjeev P. Bhavnani |
| author_sort | Christine P. Shen |
| collection | DOAJ |
| description | Background Diagnosis of shockable rhythms leading to defibrillation remains integral to improving out‐of‐hospital cardiac arrest outcomes. New machine learning techniques have emerged to diagnose arrhythmias on ECGs. In out‐of‐hospital cardiac arrest, an algorithm within an automated external defibrillator is the major determinant to deliver defibrillation. This study developed and validated the performance of a convolution neural network (CNN) to diagnose shockable arrhythmias within a novel, miniaturized automated external defibrillator. Methods and Results There were 26 464 single‐lead ECGs that comprised the study data set. ECGs of 7‐s duration were retrospectively adjudicated by 3 physician readers (N=18 total readers). After exclusions (N=1582), ECGs were divided into training (N=23 156), validation (N=721), and test data sets (N=1005). CNN performance to diagnose shockable and nonshockable rhythms was reported with area under the receiver operating characteristic curve analysis, F1, and sensitivity and specificity calculations. The duration for the CNN to output was reported with the algorithm running within the automated external defibrillator. Internal and external validation analyses included CNN performance among arrhythmias, often mistaken for shockable rhythms, and performance among ECGs modified with noise to mimic artifacts. The CNN algorithm achieved an area under the receiver operating characteristic curve of 0.995 (95% CI, 0.990–1.0), sensitivity of 98%, and specificity of 100% to diagnose shockable rhythms. The F1 scores were 0.990 and 0.995 for shockable and nonshockable rhythms, respectively. After input of a 7‐s ECG, the CNN generated an output in 383±29 ms (total time of 7.383 s). The CNN outperformed adjudicators in classifying atrial arrhythmias as nonshockable (specificity of 99.3%–98.1%) and was robust against noise artifacts (area under the receiver operating characteristic curve range, 0.871–0.999). Conclusions We demonstrate high diagnostic performance of a CNN algorithm for shockable and nonshockable rhythm arrhythmia classifications within a digitally connected automated external defibrillator. Registration URL: https://clinicaltrials.gov/ct2/show/NCT03662802; Unique identifier: NCT03662802 |
| first_indexed | 2024-03-11T19:02:26Z |
| format | Article |
| id | doaj.art-6f0ad7d79864433ea2fd58c7bd9757da |
| institution | Directory Open Access Journal |
| issn | 2047-9980 |
| language | English |
| last_indexed | 2024-03-11T19:02:26Z |
| publishDate | 2023-04-01 |
| publisher | Wiley |
| record_format | Article |
| series | Journal of the American Heart Association: Cardiovascular and Cerebrovascular Disease |
| spelling | doaj.art-6f0ad7d79864433ea2fd58c7bd9757da2023-10-10T11:37:42ZengWileyJournal of the American Heart Association: Cardiovascular and Cerebrovascular Disease2047-99802023-04-0112810.1161/JAHA.122.026974Convolution Neural Network Algorithm for Shockable Arrhythmia Classification Within a Digitally Connected Automated External DefibrillatorChristine P. Shen0Benjamin C. Freed1David P. Walter2James C. Perry3Amr F. Barakat4Ahmad Ramy A. Elashery5Kevin S. Shah6Shelby Kutty7Michael McGillion8Fu Siong Ng9Rola Khedraki10Keshav R. Nayak11John D. Rogers12Sanjeev P. Bhavnani13Division of Cardiology Healthcare Innovation Laboratory Scripps Clinic San Diego CACarnegie Mellon University Pittsburgh PAMassachusetts Institute of Technology Cambridge MAUniversity of California San Diego, Rady Children’s Hospital San Diego CAUniversity of Pittsburg Medical Center Pittsburgh PACharleston Area Medical Center Institute for Academic Medicine Charleston WVUniversity of Utah Health Sciences Center Salt Lake City UTJohns Hopkins University Baltimore MDMcMaster University Hamilton CanadaImperial College of London London United KingdomDivision of Cardiology Healthcare Innovation Laboratory Scripps Clinic San Diego CADivision of Interventional Cardiology Scripps Mercy Hospital San Diego CADivision of Cardiology Healthcare Innovation Laboratory Scripps Clinic San Diego CADivision of Cardiology Healthcare Innovation Laboratory Scripps Clinic San Diego CABackground Diagnosis of shockable rhythms leading to defibrillation remains integral to improving out‐of‐hospital cardiac arrest outcomes. New machine learning techniques have emerged to diagnose arrhythmias on ECGs. In out‐of‐hospital cardiac arrest, an algorithm within an automated external defibrillator is the major determinant to deliver defibrillation. This study developed and validated the performance of a convolution neural network (CNN) to diagnose shockable arrhythmias within a novel, miniaturized automated external defibrillator. Methods and Results There were 26 464 single‐lead ECGs that comprised the study data set. ECGs of 7‐s duration were retrospectively adjudicated by 3 physician readers (N=18 total readers). After exclusions (N=1582), ECGs were divided into training (N=23 156), validation (N=721), and test data sets (N=1005). CNN performance to diagnose shockable and nonshockable rhythms was reported with area under the receiver operating characteristic curve analysis, F1, and sensitivity and specificity calculations. The duration for the CNN to output was reported with the algorithm running within the automated external defibrillator. Internal and external validation analyses included CNN performance among arrhythmias, often mistaken for shockable rhythms, and performance among ECGs modified with noise to mimic artifacts. The CNN algorithm achieved an area under the receiver operating characteristic curve of 0.995 (95% CI, 0.990–1.0), sensitivity of 98%, and specificity of 100% to diagnose shockable rhythms. The F1 scores were 0.990 and 0.995 for shockable and nonshockable rhythms, respectively. After input of a 7‐s ECG, the CNN generated an output in 383±29 ms (total time of 7.383 s). The CNN outperformed adjudicators in classifying atrial arrhythmias as nonshockable (specificity of 99.3%–98.1%) and was robust against noise artifacts (area under the receiver operating characteristic curve range, 0.871–0.999). Conclusions We demonstrate high diagnostic performance of a CNN algorithm for shockable and nonshockable rhythm arrhythmia classifications within a digitally connected automated external defibrillator. Registration URL: https://clinicaltrials.gov/ct2/show/NCT03662802; Unique identifier: NCT03662802https://www.ahajournals.org/doi/10.1161/JAHA.122.026974automated external defibrillatorconvolution neural networkECGmachine learningventricular arrhythmias |
| spellingShingle | Christine P. Shen Benjamin C. Freed David P. Walter James C. Perry Amr F. Barakat Ahmad Ramy A. Elashery Kevin S. Shah Shelby Kutty Michael McGillion Fu Siong Ng Rola Khedraki Keshav R. Nayak John D. Rogers Sanjeev P. Bhavnani Convolution Neural Network Algorithm for Shockable Arrhythmia Classification Within a Digitally Connected Automated External Defibrillator Journal of the American Heart Association: Cardiovascular and Cerebrovascular Disease automated external defibrillator convolution neural network ECG machine learning ventricular arrhythmias |
| title | Convolution Neural Network Algorithm for Shockable Arrhythmia Classification Within a Digitally Connected Automated External Defibrillator |
| title_full | Convolution Neural Network Algorithm for Shockable Arrhythmia Classification Within a Digitally Connected Automated External Defibrillator |
| title_fullStr | Convolution Neural Network Algorithm for Shockable Arrhythmia Classification Within a Digitally Connected Automated External Defibrillator |
| title_full_unstemmed | Convolution Neural Network Algorithm for Shockable Arrhythmia Classification Within a Digitally Connected Automated External Defibrillator |
| title_short | Convolution Neural Network Algorithm for Shockable Arrhythmia Classification Within a Digitally Connected Automated External Defibrillator |
| title_sort | convolution neural network algorithm for shockable arrhythmia classification within a digitally connected automated external defibrillator |
| topic | automated external defibrillator convolution neural network ECG machine learning ventricular arrhythmias |
| url | https://www.ahajournals.org/doi/10.1161/JAHA.122.026974 |
| work_keys_str_mv | AT christinepshen convolutionneuralnetworkalgorithmforshockablearrhythmiaclassificationwithinadigitallyconnectedautomatedexternaldefibrillator AT benjamincfreed convolutionneuralnetworkalgorithmforshockablearrhythmiaclassificationwithinadigitallyconnectedautomatedexternaldefibrillator AT davidpwalter convolutionneuralnetworkalgorithmforshockablearrhythmiaclassificationwithinadigitallyconnectedautomatedexternaldefibrillator AT jamescperry convolutionneuralnetworkalgorithmforshockablearrhythmiaclassificationwithinadigitallyconnectedautomatedexternaldefibrillator AT amrfbarakat convolutionneuralnetworkalgorithmforshockablearrhythmiaclassificationwithinadigitallyconnectedautomatedexternaldefibrillator AT ahmadramyaelashery convolutionneuralnetworkalgorithmforshockablearrhythmiaclassificationwithinadigitallyconnectedautomatedexternaldefibrillator AT kevinsshah convolutionneuralnetworkalgorithmforshockablearrhythmiaclassificationwithinadigitallyconnectedautomatedexternaldefibrillator AT shelbykutty convolutionneuralnetworkalgorithmforshockablearrhythmiaclassificationwithinadigitallyconnectedautomatedexternaldefibrillator AT michaelmcgillion convolutionneuralnetworkalgorithmforshockablearrhythmiaclassificationwithinadigitallyconnectedautomatedexternaldefibrillator AT fusiongng convolutionneuralnetworkalgorithmforshockablearrhythmiaclassificationwithinadigitallyconnectedautomatedexternaldefibrillator AT rolakhedraki convolutionneuralnetworkalgorithmforshockablearrhythmiaclassificationwithinadigitallyconnectedautomatedexternaldefibrillator AT keshavrnayak convolutionneuralnetworkalgorithmforshockablearrhythmiaclassificationwithinadigitallyconnectedautomatedexternaldefibrillator AT johndrogers convolutionneuralnetworkalgorithmforshockablearrhythmiaclassificationwithinadigitallyconnectedautomatedexternaldefibrillator AT sanjeevpbhavnani convolutionneuralnetworkalgorithmforshockablearrhythmiaclassificationwithinadigitallyconnectedautomatedexternaldefibrillator |