Transfer learning-based electrocardiogram classification using wavelet scattered features
Background: The abnormalities in the heart rhythm result in various cardiac issues affecting the normal functioning of the heart. Early diagnosis helps prevent serious outcomes and to treat them effectively. This work focuses on classifying the various abnormalities with the changes in the heart rhy...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wolters Kluwer Medknow Publications
2023-01-01
|
| Series: | Biomedical and Biotechnology Research Journal |
| Subjects: | |
| Online Access: | http://www.bmbtrj.org/article.asp?issn=2588-9834;year=2023;volume=7;issue=1;spage=52;epage=59;aulast=Sabeenian |
| _version_ | 1797866691481305088 |
|---|---|
| author | R S Sabeenian K K Sree Janani |
| author_facet | R S Sabeenian K K Sree Janani |
| author_sort | R S Sabeenian |
| collection | DOAJ |
| description | Background: The abnormalities in the heart rhythm result in various cardiac issues affecting the normal functioning of the heart. Early diagnosis helps prevent serious outcomes and to treat them effectively. This work focuses on classifying the various abnormalities with the changes in the heart rhythm and demographic data. The pretrained convolution neural network models classify the wavelet scattered data of different arrhythmic electrocardiograms (ECGs). Methods: The ECG signals of different anomalies from the PhysioNet database are re-sampled and segmented. The sampling is done using the linear interpolation method, which estimates values between the sample points based on nearby data points. The inter-dependence variances among the data points were extracted using wavelet scattering. The one-dimensional (1D) signal data are converted into 2D scalogram images using continuous wavelet transform. Pretrained deep learning models are used to extract features from the scalogram images and classify using a support vector machine classifier. The classification results are analyzed using various performance metrics such as precision, specificity, recall, F-measure, and accuracy. The relationship between the model performance and network depth and learnables is analyzed. Results: The classification results show that the ResNet18 achieves higher accuracy of 98.81% for raw data and 97.05% for wavelet scattered data. No dependency exists between the model depth, network parameters, and performance. The ResNet18 model achieves higher precision, recall, specificity, and F-measure values of 96.49%, 96.42%, 98.24%, and 96.45%, respectively, for wavelet scattered data. Conclusions: The ResNet18 achieves generalized results in classifying dimensionality-reduced data with reduced computational cost and high accuracy. The DenseNet model achieves higher performance metrics for raw data, whereas the ResNet18 model achieves higher performance metrics for wavelet scattered data. |
| first_indexed | 2024-04-09T23:29:26Z |
| format | Article |
| id | doaj.art-2b0b1ac7726e4528b1b1216c8e30d21b |
| institution | Directory Open Access Journal |
| issn | 2588-9834 2588-9842 |
| language | English |
| last_indexed | 2024-04-09T23:29:26Z |
| publishDate | 2023-01-01 |
| publisher | Wolters Kluwer Medknow Publications |
| record_format | Article |
| series | Biomedical and Biotechnology Research Journal |
| spelling | doaj.art-2b0b1ac7726e4528b1b1216c8e30d21b2023-03-21T07:28:24ZengWolters Kluwer Medknow PublicationsBiomedical and Biotechnology Research Journal2588-98342588-98422023-01-0171525910.4103/bbrj.bbrj_341_22Transfer learning-based electrocardiogram classification using wavelet scattered featuresR S SabeenianK K Sree JananiBackground: The abnormalities in the heart rhythm result in various cardiac issues affecting the normal functioning of the heart. Early diagnosis helps prevent serious outcomes and to treat them effectively. This work focuses on classifying the various abnormalities with the changes in the heart rhythm and demographic data. The pretrained convolution neural network models classify the wavelet scattered data of different arrhythmic electrocardiograms (ECGs). Methods: The ECG signals of different anomalies from the PhysioNet database are re-sampled and segmented. The sampling is done using the linear interpolation method, which estimates values between the sample points based on nearby data points. The inter-dependence variances among the data points were extracted using wavelet scattering. The one-dimensional (1D) signal data are converted into 2D scalogram images using continuous wavelet transform. Pretrained deep learning models are used to extract features from the scalogram images and classify using a support vector machine classifier. The classification results are analyzed using various performance metrics such as precision, specificity, recall, F-measure, and accuracy. The relationship between the model performance and network depth and learnables is analyzed. Results: The classification results show that the ResNet18 achieves higher accuracy of 98.81% for raw data and 97.05% for wavelet scattered data. No dependency exists between the model depth, network parameters, and performance. The ResNet18 model achieves higher precision, recall, specificity, and F-measure values of 96.49%, 96.42%, 98.24%, and 96.45%, respectively, for wavelet scattered data. Conclusions: The ResNet18 achieves generalized results in classifying dimensionality-reduced data with reduced computational cost and high accuracy. The DenseNet model achieves higher performance metrics for raw data, whereas the ResNet18 model achieves higher performance metrics for wavelet scattered data.http://www.bmbtrj.org/article.asp?issn=2588-9834;year=2023;volume=7;issue=1;spage=52;epage=59;aulast=Sabeenianarrhythmiaelectrocardiogramscalogramtransfer learningwavelet scattering |
| spellingShingle | R S Sabeenian K K Sree Janani Transfer learning-based electrocardiogram classification using wavelet scattered features Biomedical and Biotechnology Research Journal arrhythmia electrocardiogram scalogram transfer learning wavelet scattering |
| title | Transfer learning-based electrocardiogram classification using wavelet scattered features |
| title_full | Transfer learning-based electrocardiogram classification using wavelet scattered features |
| title_fullStr | Transfer learning-based electrocardiogram classification using wavelet scattered features |
| title_full_unstemmed | Transfer learning-based electrocardiogram classification using wavelet scattered features |
| title_short | Transfer learning-based electrocardiogram classification using wavelet scattered features |
| title_sort | transfer learning based electrocardiogram classification using wavelet scattered features |
| topic | arrhythmia electrocardiogram scalogram transfer learning wavelet scattering |
| url | http://www.bmbtrj.org/article.asp?issn=2588-9834;year=2023;volume=7;issue=1;spage=52;epage=59;aulast=Sabeenian |
| work_keys_str_mv | AT rssabeenian transferlearningbasedelectrocardiogramclassificationusingwaveletscatteredfeatures AT kksreejanani transferlearningbasedelectrocardiogramclassificationusingwaveletscatteredfeatures |