ECGTransForm: empowering adaptive ECG arrhythmia classification framework with bidirectional transformer
Cardiac arrhythmias, deviations from the normal rhythmic beating of the heart, are subtle yet critical indicators of potential cardiac challenges. Efficiently diagnosing them requires intricate understanding and representation of both spatial and temporal features present in Electrocardiogram (ECG)...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Journal Article |
| Language: | English |
| Published: |
2023
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/171854 |
| _version_ | 1826120399347253248 |
|---|---|
| author | Eldele, Emadeldeen El-Ghaish, Hany |
| author2 | School of Computer Science and Engineering |
| author_facet | School of Computer Science and Engineering Eldele, Emadeldeen El-Ghaish, Hany |
| author_sort | Eldele, Emadeldeen |
| collection | NTU |
| description | Cardiac arrhythmias, deviations from the normal rhythmic beating of the heart, are subtle yet critical indicators of potential cardiac challenges. Efficiently diagnosing them requires intricate understanding and representation of both spatial and temporal features present in Electrocardiogram (ECG) signals. This paper introduces \abb, a deep learning framework tailored for ECG arrhythmia classification. By embedding a novel Bidirectional Transformer (BiTrans) mechanism, our model comprehensively captures temporal dependencies from both antecedent and subsequent contexts. This is further augmented with Multi-scale Convolutions and a Channel Recalibration Module, ensuring a robust spatial feature extraction across various granularities. We also introduce a Context-Aware Loss (CAL) that addresses the class imbalance challenge inherent in ECG datasets by dynamically adjusting weights based on class representation. Extensive experiments reveal that \abb outperforms contemporary models, proving its efficacy in extracting meaningful features for arrhythmia diagnosis. Our work offers a significant step towards enhancing the accuracy and efficiency of automated ECG-based cardiac diagnoses, with potential implications for broader cardiac care applications. The source code is available at https://github.com/emadeldeen24/ECGTransForm. |
| first_indexed | 2024-10-01T05:15:49Z |
| format | Journal Article |
| id | ntu-10356/171854 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T05:15:49Z |
| publishDate | 2023 |
| record_format | dspace |
| spelling | ntu-10356/1718542023-11-17T15:35:51Z ECGTransForm: empowering adaptive ECG arrhythmia classification framework with bidirectional transformer Eldele, Emadeldeen El-Ghaish, Hany School of Computer Science and Engineering Engineering::Bioengineering ECG Arrhythmia Cardiac arrhythmias, deviations from the normal rhythmic beating of the heart, are subtle yet critical indicators of potential cardiac challenges. Efficiently diagnosing them requires intricate understanding and representation of both spatial and temporal features present in Electrocardiogram (ECG) signals. This paper introduces \abb, a deep learning framework tailored for ECG arrhythmia classification. By embedding a novel Bidirectional Transformer (BiTrans) mechanism, our model comprehensively captures temporal dependencies from both antecedent and subsequent contexts. This is further augmented with Multi-scale Convolutions and a Channel Recalibration Module, ensuring a robust spatial feature extraction across various granularities. We also introduce a Context-Aware Loss (CAL) that addresses the class imbalance challenge inherent in ECG datasets by dynamically adjusting weights based on class representation. Extensive experiments reveal that \abb outperforms contemporary models, proving its efficacy in extracting meaningful features for arrhythmia diagnosis. Our work offers a significant step towards enhancing the accuracy and efficiency of automated ECG-based cardiac diagnoses, with potential implications for broader cardiac care applications. The source code is available at https://github.com/emadeldeen24/ECGTransForm. Submitted/Accepted version 2023-11-16T01:15:08Z 2023-11-16T01:15:08Z 2024 Journal Article Eldele, E. & El-Ghaish, H. (2024). ECGTransForm: empowering adaptive ECG arrhythmia classification framework with bidirectional transformer. Biomedical Signal Processing and Control, 89, 105714-. https://dx.doi.org/10.1016/j.bspc.2023.105714 1746-8094 https://hdl.handle.net/10356/171854 10.1016/j.bspc.2023.105714 89 105714 en Biomedical Signal Processing and Control © 2023 Elsevier Ltd. All rights reserved. This article may be downloaded for personal use only. Any other use requires prior permission of the copyright holder. The Version of Record is available online at http://doi.org/10.1016/j.bspc.2023.105714. application/pdf |
| spellingShingle | Engineering::Bioengineering ECG Arrhythmia Eldele, Emadeldeen El-Ghaish, Hany ECGTransForm: empowering adaptive ECG arrhythmia classification framework with bidirectional transformer |
| title | ECGTransForm: empowering adaptive ECG arrhythmia classification framework with bidirectional transformer |
| title_full | ECGTransForm: empowering adaptive ECG arrhythmia classification framework with bidirectional transformer |
| title_fullStr | ECGTransForm: empowering adaptive ECG arrhythmia classification framework with bidirectional transformer |
| title_full_unstemmed | ECGTransForm: empowering adaptive ECG arrhythmia classification framework with bidirectional transformer |
| title_short | ECGTransForm: empowering adaptive ECG arrhythmia classification framework with bidirectional transformer |
| title_sort | ecgtransform empowering adaptive ecg arrhythmia classification framework with bidirectional transformer |
| topic | Engineering::Bioengineering ECG Arrhythmia |
| url | https://hdl.handle.net/10356/171854 |
| work_keys_str_mv | AT eldeleemadeldeen ecgtransformempoweringadaptiveecgarrhythmiaclassificationframeworkwithbidirectionaltransformer AT elghaishhany ecgtransformempoweringadaptiveecgarrhythmiaclassificationframeworkwithbidirectionaltransformer |