Analyzing artificial intelligence systems for the prediction of atrial fibrillation from sinus-rhythm ECGs including demographics and feature visualization
Abstract Atrial fibrillation (AF) is an abnormal heart rhythm, asymptomatic in many cases, that causes several health problems and mortality in population. This retrospective study evaluates the ability of different AI-based models to predict future episodes of AF from electrocardiograms (ECGs) reco...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2021-11-01
|
| Series: | Scientific Reports |
| Online Access: | https://doi.org/10.1038/s41598-021-02179-1 |
| _version_ | 1830301403015282688 |
|---|---|
| author | Pietro Melzi Ruben Tolosana Alberto Cecconi Ancor Sanz-Garcia Guillermo J. Ortega Luis Jesus Jimenez-Borreguero Ruben Vera-Rodriguez |
| author_facet | Pietro Melzi Ruben Tolosana Alberto Cecconi Ancor Sanz-Garcia Guillermo J. Ortega Luis Jesus Jimenez-Borreguero Ruben Vera-Rodriguez |
| author_sort | Pietro Melzi |
| collection | DOAJ |
| description | Abstract Atrial fibrillation (AF) is an abnormal heart rhythm, asymptomatic in many cases, that causes several health problems and mortality in population. This retrospective study evaluates the ability of different AI-based models to predict future episodes of AF from electrocardiograms (ECGs) recorded during normal sinus rhythm. Patients are divided into two classes according to AF occurrence or sinus rhythm permanence along their several ECGs registry. In the constrained scenario of balancing the age distributions between classes, our best AI model predicts future episodes of AF with area under the curve (AUC) 0.79 (0.72–0.86). Multiple scenarios and age-sex-specific groups of patients are considered, achieving best performance of prediction for males older than 70 years. These results point out the importance of considering different demographic groups in the analysis of AF prediction, showing considerable performance gaps among them. In addition to the demographic analysis, we apply feature visualization techniques to identify the most important portions of the ECG signals in the task of AF prediction, improving this way the interpretability and understanding of the AI models. These results and the simplicity of recording ECGs during check-ups add feasibility to clinical applications of AI-based models. |
| first_indexed | 2024-12-19T08:48:35Z |
| format | Article |
| id | doaj.art-d3fe7ade268d446db2798bffbd1c1372 |
| institution | Directory Open Access Journal |
| issn | 2045-2322 |
| language | English |
| last_indexed | 2024-12-19T08:48:35Z |
| publishDate | 2021-11-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj.art-d3fe7ade268d446db2798bffbd1c13722022-12-21T20:28:47ZengNature PortfolioScientific Reports2045-23222021-11-0111111010.1038/s41598-021-02179-1Analyzing artificial intelligence systems for the prediction of atrial fibrillation from sinus-rhythm ECGs including demographics and feature visualizationPietro Melzi0Ruben Tolosana1Alberto Cecconi2Ancor Sanz-Garcia3Guillermo J. Ortega4Luis Jesus Jimenez-Borreguero5Ruben Vera-Rodriguez6Biometrics and Data Pattern Analytics Lab, Escuela Politecnica Superior, Universidad Autonoma de MadridBiometrics and Data Pattern Analytics Lab, Escuela Politecnica Superior, Universidad Autonoma de MadridInstituto de Investigacion Sanitaria del Hospital Universitario de La PrincesaInstituto de Investigacion Sanitaria del Hospital Universitario de La PrincesaScience and Technology Department, National University of QuilmesInstituto de Investigacion Sanitaria del Hospital Universitario de La PrincesaBiometrics and Data Pattern Analytics Lab, Escuela Politecnica Superior, Universidad Autonoma de MadridAbstract Atrial fibrillation (AF) is an abnormal heart rhythm, asymptomatic in many cases, that causes several health problems and mortality in population. This retrospective study evaluates the ability of different AI-based models to predict future episodes of AF from electrocardiograms (ECGs) recorded during normal sinus rhythm. Patients are divided into two classes according to AF occurrence or sinus rhythm permanence along their several ECGs registry. In the constrained scenario of balancing the age distributions between classes, our best AI model predicts future episodes of AF with area under the curve (AUC) 0.79 (0.72–0.86). Multiple scenarios and age-sex-specific groups of patients are considered, achieving best performance of prediction for males older than 70 years. These results point out the importance of considering different demographic groups in the analysis of AF prediction, showing considerable performance gaps among them. In addition to the demographic analysis, we apply feature visualization techniques to identify the most important portions of the ECG signals in the task of AF prediction, improving this way the interpretability and understanding of the AI models. These results and the simplicity of recording ECGs during check-ups add feasibility to clinical applications of AI-based models.https://doi.org/10.1038/s41598-021-02179-1 |
| spellingShingle | Pietro Melzi Ruben Tolosana Alberto Cecconi Ancor Sanz-Garcia Guillermo J. Ortega Luis Jesus Jimenez-Borreguero Ruben Vera-Rodriguez Analyzing artificial intelligence systems for the prediction of atrial fibrillation from sinus-rhythm ECGs including demographics and feature visualization Scientific Reports |
| title | Analyzing artificial intelligence systems for the prediction of atrial fibrillation from sinus-rhythm ECGs including demographics and feature visualization |
| title_full | Analyzing artificial intelligence systems for the prediction of atrial fibrillation from sinus-rhythm ECGs including demographics and feature visualization |
| title_fullStr | Analyzing artificial intelligence systems for the prediction of atrial fibrillation from sinus-rhythm ECGs including demographics and feature visualization |
| title_full_unstemmed | Analyzing artificial intelligence systems for the prediction of atrial fibrillation from sinus-rhythm ECGs including demographics and feature visualization |
| title_short | Analyzing artificial intelligence systems for the prediction of atrial fibrillation from sinus-rhythm ECGs including demographics and feature visualization |
| title_sort | analyzing artificial intelligence systems for the prediction of atrial fibrillation from sinus rhythm ecgs including demographics and feature visualization |
| url | https://doi.org/10.1038/s41598-021-02179-1 |
| work_keys_str_mv | AT pietromelzi analyzingartificialintelligencesystemsforthepredictionofatrialfibrillationfromsinusrhythmecgsincludingdemographicsandfeaturevisualization AT rubentolosana analyzingartificialintelligencesystemsforthepredictionofatrialfibrillationfromsinusrhythmecgsincludingdemographicsandfeaturevisualization AT albertocecconi analyzingartificialintelligencesystemsforthepredictionofatrialfibrillationfromsinusrhythmecgsincludingdemographicsandfeaturevisualization AT ancorsanzgarcia analyzingartificialintelligencesystemsforthepredictionofatrialfibrillationfromsinusrhythmecgsincludingdemographicsandfeaturevisualization AT guillermojortega analyzingartificialintelligencesystemsforthepredictionofatrialfibrillationfromsinusrhythmecgsincludingdemographicsandfeaturevisualization AT luisjesusjimenezborreguero analyzingartificialintelligencesystemsforthepredictionofatrialfibrillationfromsinusrhythmecgsincludingdemographicsandfeaturevisualization AT rubenverarodriguez analyzingartificialintelligencesystemsforthepredictionofatrialfibrillationfromsinusrhythmecgsincludingdemographicsandfeaturevisualization |