Classification of Bundle Branch Blocks with QRS Templates Extracted from 12-lead ECGs
Being non-invasive, cheap and widely available, the 12-lead electrocardiogram (ECG) is a standard method to assess cardiac function. Still, its reliable interpretation requires specialized knowledge and experience, rendering a second opinion valuable. We evaluated the performance of machine learning...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
De Gruyter
2021-10-01
|
| Series: | Current Directions in Biomedical Engineering |
| Subjects: | |
| Online Access: | https://doi.org/10.1515/cdbme-2021-2148 |
| _version_ | 1818509896585314304 |
|---|---|
| author | Welle Hannes Nagel Claudia Loewe Axel Mikut Ralf Dössel Olaf |
| author_facet | Welle Hannes Nagel Claudia Loewe Axel Mikut Ralf Dössel Olaf |
| author_sort | Welle Hannes |
| collection | DOAJ |
| description | Being non-invasive, cheap and widely available, the 12-lead electrocardiogram (ECG) is a standard method to assess cardiac function. Still, its reliable interpretation requires specialized knowledge and experience, rendering a second opinion valuable. We evaluated the performance of machine learning based classification of 11,705 healthy and bundle branch block 12-lead ECGs from 3 open databases. For each lead of the ECG signal, a representative QRS-complex template was extracted automatically. Principal component analysis (PCA) was applied to the concatenated, normalized and rescaled QRS signals to reduce their dimensionality. Multilayer perceptron and support-vector machine classifiers were trained using the principal components of weighted and non-weighted QRS template signals as input data. Classifiers achieved F1 scores between 0.92 and 0.96 on the test set for different input configurations. Anomaly based weighting slightly improved the performance of the classifiers. Neither class-wise PCA for feature extraction nor adding information on sex, gender and electrical heart axis to the input data yielded considerable improvement of the F1 scores. The achieved classification accuracy is similar to deep learning classifier performances and should generalize robustly to other ECG datasets. Our results suggest that this simple and well interpretable approach based on morphological signal characteristics is suitable for automatically and non-invasively identifying bundle branch block pathologies in clinical or smart electronics contexts. |
| first_indexed | 2024-12-10T22:51:41Z |
| format | Article |
| id | doaj.art-88379fe9ab8846018ded4dccdc2e8677 |
| institution | Directory Open Access Journal |
| issn | 2364-5504 |
| language | English |
| last_indexed | 2024-12-10T22:51:41Z |
| publishDate | 2021-10-01 |
| publisher | De Gruyter |
| record_format | Article |
| series | Current Directions in Biomedical Engineering |
| spelling | doaj.art-88379fe9ab8846018ded4dccdc2e86772022-12-22T01:30:23ZengDe GruyterCurrent Directions in Biomedical Engineering2364-55042021-10-017258258510.1515/cdbme-2021-2148Classification of Bundle Branch Blocks with QRS Templates Extracted from 12-lead ECGsWelle Hannes0Nagel Claudia1Loewe Axel2Mikut Ralf3Dössel Olaf4Institute of Biomedical Engineering KIT, Fritz-Haber-Weg 1,Karlsruhe, GermanyInstitute of Biomedical Engineering KIT, Fritz-Haber-Weg 1,Karlsruhe, GermanyInstitute of Biomedical Engineering KIT, Fritz-Haber-Weg 1,Karlsruhe, GermanyInstitute for Automation and Applied Informatics KIT, Hermann-von-Helmholtz-Platz 1,Eggenstein-Leopoldshafen, GermanyInstitute of Biomedical Engineering KIT, Fritz-Haber-Weg 1,Karlsruhe, GermanyBeing non-invasive, cheap and widely available, the 12-lead electrocardiogram (ECG) is a standard method to assess cardiac function. Still, its reliable interpretation requires specialized knowledge and experience, rendering a second opinion valuable. We evaluated the performance of machine learning based classification of 11,705 healthy and bundle branch block 12-lead ECGs from 3 open databases. For each lead of the ECG signal, a representative QRS-complex template was extracted automatically. Principal component analysis (PCA) was applied to the concatenated, normalized and rescaled QRS signals to reduce their dimensionality. Multilayer perceptron and support-vector machine classifiers were trained using the principal components of weighted and non-weighted QRS template signals as input data. Classifiers achieved F1 scores between 0.92 and 0.96 on the test set for different input configurations. Anomaly based weighting slightly improved the performance of the classifiers. Neither class-wise PCA for feature extraction nor adding information on sex, gender and electrical heart axis to the input data yielded considerable improvement of the F1 scores. The achieved classification accuracy is similar to deep learning classifier performances and should generalize robustly to other ECG datasets. Our results suggest that this simple and well interpretable approach based on morphological signal characteristics is suitable for automatically and non-invasively identifying bundle branch block pathologies in clinical or smart electronics contexts.https://doi.org/10.1515/cdbme-2021-214812 lead ecgqrsclassificationbundle branch blockpreprocessingmachine learning |
| spellingShingle | Welle Hannes Nagel Claudia Loewe Axel Mikut Ralf Dössel Olaf Classification of Bundle Branch Blocks with QRS Templates Extracted from 12-lead ECGs Current Directions in Biomedical Engineering 12 lead ecg qrs classification bundle branch block preprocessing machine learning |
| title | Classification of Bundle Branch Blocks with QRS Templates Extracted from 12-lead ECGs |
| title_full | Classification of Bundle Branch Blocks with QRS Templates Extracted from 12-lead ECGs |
| title_fullStr | Classification of Bundle Branch Blocks with QRS Templates Extracted from 12-lead ECGs |
| title_full_unstemmed | Classification of Bundle Branch Blocks with QRS Templates Extracted from 12-lead ECGs |
| title_short | Classification of Bundle Branch Blocks with QRS Templates Extracted from 12-lead ECGs |
| title_sort | classification of bundle branch blocks with qrs templates extracted from 12 lead ecgs |
| topic | 12 lead ecg qrs classification bundle branch block preprocessing machine learning |
| url | https://doi.org/10.1515/cdbme-2021-2148 |
| work_keys_str_mv | AT wellehannes classificationofbundlebranchblockswithqrstemplatesextractedfrom12leadecgs AT nagelclaudia classificationofbundlebranchblockswithqrstemplatesextractedfrom12leadecgs AT loeweaxel classificationofbundlebranchblockswithqrstemplatesextractedfrom12leadecgs AT mikutralf classificationofbundlebranchblockswithqrstemplatesextractedfrom12leadecgs AT dosselolaf classificationofbundlebranchblockswithqrstemplatesextractedfrom12leadecgs |