Pragmatic screening for heart failure in the general population using an electrocardiogram‐based neural network
Abstract Aims We aim to develop a pragmatic screening tool for heart failure at the general population level. Methods and results This study was conducted within the Hamburg‐City‐Health‐Study, an ongoing, prospective, observational study enrolling randomly selected inhabitants of the city of Hamburg...
| Main Authors: | , , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2023-04-01
|
| Series: | ESC Heart Failure |
| Subjects: | |
| Online Access: | https://doi.org/10.1002/ehf2.14263 |
| _version_ | 1797857618795954176 |
|---|---|
| author | Kishore Surendra Sylvia Nürnberg Jan P. Bremer Marius S. Knorr Frank Ückert Jan Per Wenzel Ramona Bei der Kellen Dirk Westermann Renate B. Schnabel Raphael Twerenbold Christina Magnussen Paulus Kirchhof Stefan Blankenberg Johannes Neumann Benedikt Schrage |
| author_facet | Kishore Surendra Sylvia Nürnberg Jan P. Bremer Marius S. Knorr Frank Ückert Jan Per Wenzel Ramona Bei der Kellen Dirk Westermann Renate B. Schnabel Raphael Twerenbold Christina Magnussen Paulus Kirchhof Stefan Blankenberg Johannes Neumann Benedikt Schrage |
| author_sort | Kishore Surendra |
| collection | DOAJ |
| description | Abstract Aims We aim to develop a pragmatic screening tool for heart failure at the general population level. Methods and results This study was conducted within the Hamburg‐City‐Health‐Study, an ongoing, prospective, observational study enrolling randomly selected inhabitants of the city of Hamburg aged 45–75 years. Heart failure was diagnosed per current guidelines. Using only digital electrocardiograms (ECGs), a convolutional neural network (CNN) was built to discriminate participants with and without heart failure. As comparisons, known risk variables for heart failure were fitted into a logistic regression model and a random forest classifier. Of the 5299 individuals included into this study, 318 individuals (6.0%) had heart failure. Using only the digital ECGs instead of several risk variables as an input, the CNN provided a comparable predictive accuracy for heart failure versus the logistic regression model and the random forest classifier [area under the curve (AUC) of 0.75, a sensitivity of 0.67 and a specificity of 0.69 for the CNN; AUC 0.77, a sensitivity of 0.63 and a specificity of 0.76 for the logistic regression; AUC 0.79, a sensitivity of 0.67 and a specificity of 0.72 for the random forest classifier]. Conclusions Using a CNN build on digital ECGs only and requiring no additional input, we derived a screening tool for heart failure in the general population. This could be perfectly embedded into clinical routine of general practitioners, as it builds on an already established diagnostic tool and does not require additional, time‐consuming input. This could help to alleviate the underdiagnosis of heart failure. |
| first_indexed | 2024-04-09T20:59:38Z |
| format | Article |
| id | doaj.art-5df39e51c5544b8ab6ddca88ae32af12 |
| institution | Directory Open Access Journal |
| issn | 2055-5822 |
| language | English |
| last_indexed | 2024-04-09T20:59:38Z |
| publishDate | 2023-04-01 |
| publisher | Wiley |
| record_format | Article |
| series | ESC Heart Failure |
| spelling | doaj.art-5df39e51c5544b8ab6ddca88ae32af122023-03-29T11:45:21ZengWileyESC Heart Failure2055-58222023-04-0110297598410.1002/ehf2.14263Pragmatic screening for heart failure in the general population using an electrocardiogram‐based neural networkKishore Surendra0Sylvia Nürnberg1Jan P. Bremer2Marius S. Knorr3Frank Ückert4Jan Per Wenzel5Ramona Bei der Kellen6Dirk Westermann7Renate B. Schnabel8Raphael Twerenbold9Christina Magnussen10Paulus Kirchhof11Stefan Blankenberg12Johannes Neumann13Benedikt Schrage14Department of Cardiology University Heart and Vascular Center Hamburg Hamburg GermanyInstitute of Applied Medical Informatics University Hospital Hamburg‐Eppendorf Hamburg GermanyDepartment of Cardiology University Heart and Vascular Center Hamburg Hamburg GermanyDepartment of Cardiology University Heart and Vascular Center Hamburg Hamburg GermanyInstitute of Applied Medical Informatics University Hospital Hamburg‐Eppendorf Hamburg GermanyDepartment of Cardiology University Heart and Vascular Center Hamburg Hamburg GermanyDepartment of Cardiology University Heart and Vascular Center Hamburg Hamburg GermanyDepartment of Cardiology University Heart and Vascular Center Hamburg Hamburg GermanyDepartment of Cardiology University Heart and Vascular Center Hamburg Hamburg GermanyDepartment of Cardiology University Heart and Vascular Center Hamburg Hamburg GermanyDepartment of Cardiology University Heart and Vascular Center Hamburg Hamburg GermanyDepartment of Cardiology University Heart and Vascular Center Hamburg Hamburg GermanyDepartment of Cardiology University Heart and Vascular Center Hamburg Hamburg GermanyDepartment of Cardiology University Heart and Vascular Center Hamburg Hamburg GermanyDepartment of Cardiology University Heart and Vascular Center Hamburg Hamburg GermanyAbstract Aims We aim to develop a pragmatic screening tool for heart failure at the general population level. Methods and results This study was conducted within the Hamburg‐City‐Health‐Study, an ongoing, prospective, observational study enrolling randomly selected inhabitants of the city of Hamburg aged 45–75 years. Heart failure was diagnosed per current guidelines. Using only digital electrocardiograms (ECGs), a convolutional neural network (CNN) was built to discriminate participants with and without heart failure. As comparisons, known risk variables for heart failure were fitted into a logistic regression model and a random forest classifier. Of the 5299 individuals included into this study, 318 individuals (6.0%) had heart failure. Using only the digital ECGs instead of several risk variables as an input, the CNN provided a comparable predictive accuracy for heart failure versus the logistic regression model and the random forest classifier [area under the curve (AUC) of 0.75, a sensitivity of 0.67 and a specificity of 0.69 for the CNN; AUC 0.77, a sensitivity of 0.63 and a specificity of 0.76 for the logistic regression; AUC 0.79, a sensitivity of 0.67 and a specificity of 0.72 for the random forest classifier]. Conclusions Using a CNN build on digital ECGs only and requiring no additional input, we derived a screening tool for heart failure in the general population. This could be perfectly embedded into clinical routine of general practitioners, as it builds on an already established diagnostic tool and does not require additional, time‐consuming input. This could help to alleviate the underdiagnosis of heart failure.https://doi.org/10.1002/ehf2.14263Heart failureScreeningPragmaticPopulation |
| spellingShingle | Kishore Surendra Sylvia Nürnberg Jan P. Bremer Marius S. Knorr Frank Ückert Jan Per Wenzel Ramona Bei der Kellen Dirk Westermann Renate B. Schnabel Raphael Twerenbold Christina Magnussen Paulus Kirchhof Stefan Blankenberg Johannes Neumann Benedikt Schrage Pragmatic screening for heart failure in the general population using an electrocardiogram‐based neural network ESC Heart Failure Heart failure Screening Pragmatic Population |
| title | Pragmatic screening for heart failure in the general population using an electrocardiogram‐based neural network |
| title_full | Pragmatic screening for heart failure in the general population using an electrocardiogram‐based neural network |
| title_fullStr | Pragmatic screening for heart failure in the general population using an electrocardiogram‐based neural network |
| title_full_unstemmed | Pragmatic screening for heart failure in the general population using an electrocardiogram‐based neural network |
| title_short | Pragmatic screening for heart failure in the general population using an electrocardiogram‐based neural network |
| title_sort | pragmatic screening for heart failure in the general population using an electrocardiogram based neural network |
| topic | Heart failure Screening Pragmatic Population |
| url | https://doi.org/10.1002/ehf2.14263 |
| work_keys_str_mv | AT kishoresurendra pragmaticscreeningforheartfailureinthegeneralpopulationusinganelectrocardiogrambasedneuralnetwork AT sylvianurnberg pragmaticscreeningforheartfailureinthegeneralpopulationusinganelectrocardiogrambasedneuralnetwork AT janpbremer pragmaticscreeningforheartfailureinthegeneralpopulationusinganelectrocardiogrambasedneuralnetwork AT mariussknorr pragmaticscreeningforheartfailureinthegeneralpopulationusinganelectrocardiogrambasedneuralnetwork AT frankuckert pragmaticscreeningforheartfailureinthegeneralpopulationusinganelectrocardiogrambasedneuralnetwork AT janperwenzel pragmaticscreeningforheartfailureinthegeneralpopulationusinganelectrocardiogrambasedneuralnetwork AT ramonabeiderkellen pragmaticscreeningforheartfailureinthegeneralpopulationusinganelectrocardiogrambasedneuralnetwork AT dirkwestermann pragmaticscreeningforheartfailureinthegeneralpopulationusinganelectrocardiogrambasedneuralnetwork AT renatebschnabel pragmaticscreeningforheartfailureinthegeneralpopulationusinganelectrocardiogrambasedneuralnetwork AT raphaeltwerenbold pragmaticscreeningforheartfailureinthegeneralpopulationusinganelectrocardiogrambasedneuralnetwork AT christinamagnussen pragmaticscreeningforheartfailureinthegeneralpopulationusinganelectrocardiogrambasedneuralnetwork AT pauluskirchhof pragmaticscreeningforheartfailureinthegeneralpopulationusinganelectrocardiogrambasedneuralnetwork AT stefanblankenberg pragmaticscreeningforheartfailureinthegeneralpopulationusinganelectrocardiogrambasedneuralnetwork AT johannesneumann pragmaticscreeningforheartfailureinthegeneralpopulationusinganelectrocardiogrambasedneuralnetwork AT benediktschrage pragmaticscreeningforheartfailureinthegeneralpopulationusinganelectrocardiogrambasedneuralnetwork |