Multiparametric analysis of geometric features of fibrotic textures leading to cardiac arrhythmias
Abstract One of the important questions in cardiac electrophysiology is to characterise the arrhythmogenic substrate; for example, from the texture of the cardiac fibrosis, which is considered one of the major arrhythmogenic conditions. In this paper, we perform an extensive in silico study of the r...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2021-10-01
|
| Series: | Scientific Reports |
| Online Access: | https://doi.org/10.1038/s41598-021-00606-x |
| _version_ | 1819118273470398464 |
|---|---|
| author | T. Nezlobinsky A. Okenov A. V. Panfilov |
| author_facet | T. Nezlobinsky A. Okenov A. V. Panfilov |
| author_sort | T. Nezlobinsky |
| collection | DOAJ |
| description | Abstract One of the important questions in cardiac electrophysiology is to characterise the arrhythmogenic substrate; for example, from the texture of the cardiac fibrosis, which is considered one of the major arrhythmogenic conditions. In this paper, we perform an extensive in silico study of the relationships between various local geometric characteristics of fibrosis on the onset of cardiac arrhythmias. In order to define which texture characteristics have better predictive value, we induce arrhythmias by external stimulation, selecting 4363 textures in which arrhythmia can be induced and also selecting 4363 non-arrhythmogenic textures. For each texture, we determine such characteristics as cluster area, solidity, mean distance, local density and zig-zag propagation path, and compare them in arrhythmogenic and non-arrhythmogenic cases. Our study shows that geometrical characteristics, such as cluster area or solidity, turn out to be the most important for prediction of the arrhythmogenic textures. Overall, we were able to achieve an accuracy of 67% for the arrhythmogenic texture-classification problem. However, the accuracy of predictions depends on the size of the region chosen for the analysis. The optimal size for the local areas of the tissue was of the order of 0.28 of the wavelength of the arrhythmia. We discuss further developments and possible applications of this method for characterising the substrate of arrhythmias in fibrotic textures. |
| first_indexed | 2024-12-22T05:46:15Z |
| format | Article |
| id | doaj.art-9aaf0016b8bf452991a7de42d0ab743a |
| institution | Directory Open Access Journal |
| issn | 2045-2322 |
| language | English |
| last_indexed | 2024-12-22T05:46:15Z |
| publishDate | 2021-10-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj.art-9aaf0016b8bf452991a7de42d0ab743a2022-12-21T18:37:01ZengNature PortfolioScientific Reports2045-23222021-10-0111111310.1038/s41598-021-00606-xMultiparametric analysis of geometric features of fibrotic textures leading to cardiac arrhythmiasT. Nezlobinsky0A. Okenov1A. V. Panfilov2Department of Physics and Astronomy, Ghent UniversityDepartment of Physics and Astronomy, Ghent UniversityDepartment of Physics and Astronomy, Ghent UniversityAbstract One of the important questions in cardiac electrophysiology is to characterise the arrhythmogenic substrate; for example, from the texture of the cardiac fibrosis, which is considered one of the major arrhythmogenic conditions. In this paper, we perform an extensive in silico study of the relationships between various local geometric characteristics of fibrosis on the onset of cardiac arrhythmias. In order to define which texture characteristics have better predictive value, we induce arrhythmias by external stimulation, selecting 4363 textures in which arrhythmia can be induced and also selecting 4363 non-arrhythmogenic textures. For each texture, we determine such characteristics as cluster area, solidity, mean distance, local density and zig-zag propagation path, and compare them in arrhythmogenic and non-arrhythmogenic cases. Our study shows that geometrical characteristics, such as cluster area or solidity, turn out to be the most important for prediction of the arrhythmogenic textures. Overall, we were able to achieve an accuracy of 67% for the arrhythmogenic texture-classification problem. However, the accuracy of predictions depends on the size of the region chosen for the analysis. The optimal size for the local areas of the tissue was of the order of 0.28 of the wavelength of the arrhythmia. We discuss further developments and possible applications of this method for characterising the substrate of arrhythmias in fibrotic textures.https://doi.org/10.1038/s41598-021-00606-x |
| spellingShingle | T. Nezlobinsky A. Okenov A. V. Panfilov Multiparametric analysis of geometric features of fibrotic textures leading to cardiac arrhythmias Scientific Reports |
| title | Multiparametric analysis of geometric features of fibrotic textures leading to cardiac arrhythmias |
| title_full | Multiparametric analysis of geometric features of fibrotic textures leading to cardiac arrhythmias |
| title_fullStr | Multiparametric analysis of geometric features of fibrotic textures leading to cardiac arrhythmias |
| title_full_unstemmed | Multiparametric analysis of geometric features of fibrotic textures leading to cardiac arrhythmias |
| title_short | Multiparametric analysis of geometric features of fibrotic textures leading to cardiac arrhythmias |
| title_sort | multiparametric analysis of geometric features of fibrotic textures leading to cardiac arrhythmias |
| url | https://doi.org/10.1038/s41598-021-00606-x |
| work_keys_str_mv | AT tnezlobinsky multiparametricanalysisofgeometricfeaturesoffibrotictexturesleadingtocardiacarrhythmias AT aokenov multiparametricanalysisofgeometricfeaturesoffibrotictexturesleadingtocardiacarrhythmias AT avpanfilov multiparametricanalysisofgeometricfeaturesoffibrotictexturesleadingtocardiacarrhythmias |