A convolutional autoencoder framework for ECG signal analysis
Electrocardiographic (ECG) signals are used to evaluate heart activity and to identify disease-related anomalies. Reliable support systems are useful for analyzing ECG signals, for instance, in long-term data acquisition and evaluation (e.g., 24-hour holter recording) or to support physicians in rea...
Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-01-01
|
| Series: | Heliyon |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2405844024175482 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832573123041427456 |
|---|---|
| author | Ugo Lomoio Patrizia Vizza Raffaele Giancotti Salvatore Petrolo Sergio Flesca Fabiola Boccuto Pietro Hiram Guzzi Pierangelo Veltri Giuseppe Tradigo |
| author_facet | Ugo Lomoio Patrizia Vizza Raffaele Giancotti Salvatore Petrolo Sergio Flesca Fabiola Boccuto Pietro Hiram Guzzi Pierangelo Veltri Giuseppe Tradigo |
| author_sort | Ugo Lomoio |
| collection | DOAJ |
| description | Electrocardiographic (ECG) signals are used to evaluate heart activity and to identify disease-related anomalies. Reliable support systems are useful for analyzing ECG signals, for instance, in long-term data acquisition and evaluation (e.g., 24-hour holter recording) or to support physicians in reading ECGs. Analysis of time varying signals may be done by using autoencoders (AEs) deep neural networks. AE specialized for signal data, named Convolutional Autoencoder (CAE), showed the best performances in the analysis of ECG signals.This paper presents a CAE-based framework for ECG signal analysis and anomaly identification. The trained phase is performed on synthetic data signals. The trained neural network obtained is used for the detection of anomalies in ECG signals. The trained framework has been tested on 12 lead ECG signals on a benchmark dataset and applied in scenarios where anomalies are related to cardiological risks and pathologies. The results show interesting results in automatic anomaly detection to support physicians in the decision process. The results show that the CAE-based framework is able to identify anomalies in ECG signals with a ROC AUC of 97.82% on simulated test set and a ROC AUC of 99.75% using a real test set. Finally, the proposed method has been enriched by means of reconstruction error based explainability modules and time-windows based preprocessing modules. Explainability results have been validated using abnormalities annotated by a cardiologist as ground truth and compared with explainations results. System with both code and data, is available at https://github.com/UgoLomoio/ECG_DSS_CAE. |
| format | Article |
| id | doaj-art-05c455623e0d4dfa8836cf7658167276 |
| institution | Kabale University |
| issn | 2405-8440 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Heliyon |
| spelling | doaj-art-05c455623e0d4dfa8836cf76581672762025-02-02T05:27:49ZengElsevierHeliyon2405-84402025-01-01112e41517A convolutional autoencoder framework for ECG signal analysisUgo Lomoio0Patrizia Vizza1Raffaele Giancotti2Salvatore Petrolo3Sergio Flesca4Fabiola Boccuto5Pietro Hiram Guzzi6Pierangelo Veltri7Giuseppe Tradigo8Department of Surgical and Medical Sciences, University of Catanzaro, Catanzaro, ItalyDepartment of Surgical and Medical Sciences, University of Catanzaro, Catanzaro, ItalyDepartment of Surgical and Medical Sciences, University of Catanzaro, Catanzaro, ItalyDIMES, University of Calabria, Rende, ItalyDIMES, University of Calabria, Rende, ItalyDivision of Cardiology, Department of Surgical and Medical Sciences, University of Catanzaro, Catanzaro, ItalyDepartment of Surgical and Medical Sciences, University of Catanzaro, Catanzaro, Italy; Corresponding author.DIMES, University of Calabria, Rende, ItalySMARTEST Laboratory, E-Campus University, Novedrate, ItalyElectrocardiographic (ECG) signals are used to evaluate heart activity and to identify disease-related anomalies. Reliable support systems are useful for analyzing ECG signals, for instance, in long-term data acquisition and evaluation (e.g., 24-hour holter recording) or to support physicians in reading ECGs. Analysis of time varying signals may be done by using autoencoders (AEs) deep neural networks. AE specialized for signal data, named Convolutional Autoencoder (CAE), showed the best performances in the analysis of ECG signals.This paper presents a CAE-based framework for ECG signal analysis and anomaly identification. The trained phase is performed on synthetic data signals. The trained neural network obtained is used for the detection of anomalies in ECG signals. The trained framework has been tested on 12 lead ECG signals on a benchmark dataset and applied in scenarios where anomalies are related to cardiological risks and pathologies. The results show interesting results in automatic anomaly detection to support physicians in the decision process. The results show that the CAE-based framework is able to identify anomalies in ECG signals with a ROC AUC of 97.82% on simulated test set and a ROC AUC of 99.75% using a real test set. Finally, the proposed method has been enriched by means of reconstruction error based explainability modules and time-windows based preprocessing modules. Explainability results have been validated using abnormalities annotated by a cardiologist as ground truth and compared with explainations results. System with both code and data, is available at https://github.com/UgoLomoio/ECG_DSS_CAE.http://www.sciencedirect.com/science/article/pii/S2405844024175482ECGAnomaly detectionSignal annotationAutoencoderDecision support systems |
| spellingShingle | Ugo Lomoio Patrizia Vizza Raffaele Giancotti Salvatore Petrolo Sergio Flesca Fabiola Boccuto Pietro Hiram Guzzi Pierangelo Veltri Giuseppe Tradigo A convolutional autoencoder framework for ECG signal analysis Heliyon ECG Anomaly detection Signal annotation Autoencoder Decision support systems |
| title | A convolutional autoencoder framework for ECG signal analysis |
| title_full | A convolutional autoencoder framework for ECG signal analysis |
| title_fullStr | A convolutional autoencoder framework for ECG signal analysis |
| title_full_unstemmed | A convolutional autoencoder framework for ECG signal analysis |
| title_short | A convolutional autoencoder framework for ECG signal analysis |
| title_sort | convolutional autoencoder framework for ecg signal analysis |
| topic | ECG Anomaly detection Signal annotation Autoencoder Decision support systems |
| url | http://www.sciencedirect.com/science/article/pii/S2405844024175482 |
| work_keys_str_mv | AT ugolomoio aconvolutionalautoencoderframeworkforecgsignalanalysis AT patriziavizza aconvolutionalautoencoderframeworkforecgsignalanalysis AT raffaelegiancotti aconvolutionalautoencoderframeworkforecgsignalanalysis AT salvatorepetrolo aconvolutionalautoencoderframeworkforecgsignalanalysis AT sergioflesca aconvolutionalautoencoderframeworkforecgsignalanalysis AT fabiolaboccuto aconvolutionalautoencoderframeworkforecgsignalanalysis AT pietrohiramguzzi aconvolutionalautoencoderframeworkforecgsignalanalysis AT pierangeloveltri aconvolutionalautoencoderframeworkforecgsignalanalysis AT giuseppetradigo aconvolutionalautoencoderframeworkforecgsignalanalysis AT ugolomoio convolutionalautoencoderframeworkforecgsignalanalysis AT patriziavizza convolutionalautoencoderframeworkforecgsignalanalysis AT raffaelegiancotti convolutionalautoencoderframeworkforecgsignalanalysis AT salvatorepetrolo convolutionalautoencoderframeworkforecgsignalanalysis AT sergioflesca convolutionalautoencoderframeworkforecgsignalanalysis AT fabiolaboccuto convolutionalautoencoderframeworkforecgsignalanalysis AT pietrohiramguzzi convolutionalautoencoderframeworkforecgsignalanalysis AT pierangeloveltri convolutionalautoencoderframeworkforecgsignalanalysis AT giuseppetradigo convolutionalautoencoderframeworkforecgsignalanalysis |