Rapid Exclusion of COVID Infection With the Artificial Intelligence Electrocardiogram
(2021) In Mayo Clinic Proceedings 96(8). p.2081-2094- Abstract
Objective: To rapidly exclude severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection using artificial intelligence applied to the electrocardiogram (ECG). Methods: A global, volunteer consortium from 4 continents identified patients with ECGs obtained around the time of polymerase chain reaction–confirmed COVID-19 diagnosis and age- and sex-matched controls from the same sites. Clinical characteristics, polymerase chain reaction results, and raw electrocardiographic data were collected. A convolutional neural network was trained using 26,153 ECGs (33.2% COVID positive), validated with 3826 ECGs (33.3% positive), and tested on 7870 ECGs not included in other sets (32.7% positive). Performance under different prevalence... (More)
Objective: To rapidly exclude severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection using artificial intelligence applied to the electrocardiogram (ECG). Methods: A global, volunteer consortium from 4 continents identified patients with ECGs obtained around the time of polymerase chain reaction–confirmed COVID-19 diagnosis and age- and sex-matched controls from the same sites. Clinical characteristics, polymerase chain reaction results, and raw electrocardiographic data were collected. A convolutional neural network was trained using 26,153 ECGs (33.2% COVID positive), validated with 3826 ECGs (33.3% positive), and tested on 7870 ECGs not included in other sets (32.7% positive). Performance under different prevalence values was tested by adding control ECGs from a single high-volume site. Results: The area under the curve for detection of acute COVID-19 infection in the test group was 0.767 (95% CI, 0.756 to 0.778; sensitivity, 98%; specificity, 10%; positive predictive value, 37%; negative predictive value, 91%). To more accurately reflect a real-world population, 50,905 normal controls were added to adjust the COVID prevalence to approximately 5% (2657/58,555), resulting in an area under the curve of 0.780 (95% CI, 0.771 to 0.790) with a specificity of 12.1% and a negative predictive value of 99.2%. Conclusion: Infection with SARS-CoV-2 results in electrocardiographic changes that permit the artificial intelligence–enhanced ECG to be used as a rapid screening test with a high negative predictive value (99.2%). This may permit the development of electrocardiography-based tools to rapidly screen individuals for pandemic control.
(Less)
- author
- author collaboration
- organization
- publishing date
- 2021
- type
- Contribution to journal
- publication status
- published
- subject
- in
- Mayo Clinic Proceedings
- volume
- 96
- issue
- 8
- pages
- 14 pages
- publisher
- Mayo Clinic Proceedings
- external identifiers
-
- scopus:85111611858
- pmid:34353468
- ISSN
- 0025-6196
- DOI
- 10.1016/j.mayocp.2021.05.027
- language
- English
- LU publication?
- yes
- id
- c9889b2c-5123-4512-b4cc-cddcb731d4bc
- date added to LUP
- 2021-08-27 10:52:01
- date last changed
- 2024-07-13 17:36:27
@article{c9889b2c-5123-4512-b4cc-cddcb731d4bc,
abstract = {{<p>Objective: To rapidly exclude severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection using artificial intelligence applied to the electrocardiogram (ECG). Methods: A global, volunteer consortium from 4 continents identified patients with ECGs obtained around the time of polymerase chain reaction–confirmed COVID-19 diagnosis and age- and sex-matched controls from the same sites. Clinical characteristics, polymerase chain reaction results, and raw electrocardiographic data were collected. A convolutional neural network was trained using 26,153 ECGs (33.2% COVID positive), validated with 3826 ECGs (33.3% positive), and tested on 7870 ECGs not included in other sets (32.7% positive). Performance under different prevalence values was tested by adding control ECGs from a single high-volume site. Results: The area under the curve for detection of acute COVID-19 infection in the test group was 0.767 (95% CI, 0.756 to 0.778; sensitivity, 98%; specificity, 10%; positive predictive value, 37%; negative predictive value, 91%). To more accurately reflect a real-world population, 50,905 normal controls were added to adjust the COVID prevalence to approximately 5% (2657/58,555), resulting in an area under the curve of 0.780 (95% CI, 0.771 to 0.790) with a specificity of 12.1% and a negative predictive value of 99.2%. Conclusion: Infection with SARS-CoV-2 results in electrocardiographic changes that permit the artificial intelligence–enhanced ECG to be used as a rapid screening test with a high negative predictive value (99.2%). This may permit the development of electrocardiography-based tools to rapidly screen individuals for pandemic control.</p>}},
author = {{Attia, Zachi I. and Kapa, Suraj and Dugan, Jennifer and Pereira, Naveen and Noseworthy, Peter A. and Jimenez, Francisco Lopez and Cruz, Jessica and Carter, Rickey E. and DeSimone, Daniel C. and Signorino, John and Halamka, John and Chennaiah Gari, Nikhita R. and Madathala, Raja Sekhar and Platonov, Pyotr G. and Gul, Fahad and Janssens, Stefan P. and Narayan, Sanjiv and Upadhyay, Gaurav A. and Alenghat, Francis J. and Lahiri, Marc K. and Dujardin, Karl and Hermel, Melody and Dominic, Paari and Turk-Adawi, Karam and Asaad, Nidal and Svensson, Anneli and Fernandez-Aviles, Francisco and Esakof, Darryl D. and Bartunek, Jozef and Noheria, Amit and Sridhar, Arun R. and Lanza, Gaetano A. and Cohoon, Kevin and Padmanabhan, Deepak and Pardo Gutierrez, Jose Alberto and Sinagra, Gianfranco and Merlo, Marco and Zagari, Domenico and Rodriguez Escenaro, Brenda D. and Pahlajani, Dev B. and Loncar, Goran and Vukomanovic, Vladan and Jensen, Henrik K. and Farkouh, Michael E. and Luescher, Thomas F. and Su Ping, Carolyn Lam and Peters, Nicholas S. and Friedman, Paul A.}},
issn = {{0025-6196}},
language = {{eng}},
number = {{8}},
pages = {{2081--2094}},
publisher = {{Mayo Clinic Proceedings}},
series = {{Mayo Clinic Proceedings}},
title = {{Rapid Exclusion of COVID Infection With the Artificial Intelligence Electrocardiogram}},
url = {{http://dx.doi.org/10.1016/j.mayocp.2021.05.027}},
doi = {{10.1016/j.mayocp.2021.05.027}},
volume = {{96}},
year = {{2021}},
}