Effect of Data Augmentation on Deep-Learning-Based Segmentation of Long-Axis Cine-MRI
Automated Cardiac Magnetic Resonance segmentation serves as a crucial tool for the evaluation of cardiac function, facilitating faster clinical assessments that prove advantageous for both practitioners and patients alike. Recent studies have predominantly concentrated on delineating structures on s...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-12-01
|
| Series: | Algorithms |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1999-4893/17/1/10 |
| _version_ | 1797340216425447424 |
|---|---|
| author | François Legrand Richard Macwan Alain Lalande Lisa Métairie Thomas Decourselle |
| author_facet | François Legrand Richard Macwan Alain Lalande Lisa Métairie Thomas Decourselle |
| author_sort | François Legrand |
| collection | DOAJ |
| description | Automated Cardiac Magnetic Resonance segmentation serves as a crucial tool for the evaluation of cardiac function, facilitating faster clinical assessments that prove advantageous for both practitioners and patients alike. Recent studies have predominantly concentrated on delineating structures on short-axis orientation, placing less emphasis on long-axis representations due to the intricate nature of structures in the latter. Taking these consideration into account, we present a robust hierarchy-based augmentation strategy coupled with the compact and fast Efficient-Net (ENet) architecture for the automated segmentation of two-chamber and four-chamber Cine-MRI images. We observed an average Dice improvement of 0.99% on the two-chamber images and of 2.15% on the four-chamber images, and an average Hausdorff distance improvement of 21.3% on the two-chamber images and of 29.6% on the four-chamber images. The practical viability of our approach was validated by computing clinical metrics such as the Left Ventricular Ejection Fraction (LVEF) and left ventricular volume (LVC). We observed acceptable biases, with a +2.81% deviation on the LVEF for the two-chamber images and a +0.11% deviation for the four-chamber images. |
| first_indexed | 2024-03-08T09:59:49Z |
| format | Article |
| id | doaj.art-63c142c5f7394453bbefff65318d2de6 |
| institution | Directory Open Access Journal |
| issn | 1999-4893 |
| language | English |
| last_indexed | 2024-03-08T09:59:49Z |
| publishDate | 2023-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Algorithms |
| spelling | doaj.art-63c142c5f7394453bbefff65318d2de62024-01-29T13:41:12ZengMDPI AGAlgorithms1999-48932023-12-011711010.3390/a17010010Effect of Data Augmentation on Deep-Learning-Based Segmentation of Long-Axis Cine-MRIFrançois Legrand0Richard Macwan1Alain Lalande2Lisa Métairie3Thomas Decourselle4IFTIM, ICMUB Laboratory, UMR CNRS 6302, University of Burgundy, 21078 Dijon, FranceCASIS, 21800 Quetigny, FranceIFTIM, ICMUB Laboratory, UMR CNRS 6302, University of Burgundy, 21078 Dijon, FranceCASIS, 21800 Quetigny, FranceCASIS, 21800 Quetigny, FranceAutomated Cardiac Magnetic Resonance segmentation serves as a crucial tool for the evaluation of cardiac function, facilitating faster clinical assessments that prove advantageous for both practitioners and patients alike. Recent studies have predominantly concentrated on delineating structures on short-axis orientation, placing less emphasis on long-axis representations due to the intricate nature of structures in the latter. Taking these consideration into account, we present a robust hierarchy-based augmentation strategy coupled with the compact and fast Efficient-Net (ENet) architecture for the automated segmentation of two-chamber and four-chamber Cine-MRI images. We observed an average Dice improvement of 0.99% on the two-chamber images and of 2.15% on the four-chamber images, and an average Hausdorff distance improvement of 21.3% on the two-chamber images and of 29.6% on the four-chamber images. The practical viability of our approach was validated by computing clinical metrics such as the Left Ventricular Ejection Fraction (LVEF) and left ventricular volume (LVC). We observed acceptable biases, with a +2.81% deviation on the LVEF for the two-chamber images and a +0.11% deviation for the four-chamber images.https://www.mdpi.com/1999-4893/17/1/10cine-MRIlong axisENetdata augmentationLVEFheart segmentation |
| spellingShingle | François Legrand Richard Macwan Alain Lalande Lisa Métairie Thomas Decourselle Effect of Data Augmentation on Deep-Learning-Based Segmentation of Long-Axis Cine-MRI Algorithms cine-MRI long axis ENet data augmentation LVEF heart segmentation |
| title | Effect of Data Augmentation on Deep-Learning-Based Segmentation of Long-Axis Cine-MRI |
| title_full | Effect of Data Augmentation on Deep-Learning-Based Segmentation of Long-Axis Cine-MRI |
| title_fullStr | Effect of Data Augmentation on Deep-Learning-Based Segmentation of Long-Axis Cine-MRI |
| title_full_unstemmed | Effect of Data Augmentation on Deep-Learning-Based Segmentation of Long-Axis Cine-MRI |
| title_short | Effect of Data Augmentation on Deep-Learning-Based Segmentation of Long-Axis Cine-MRI |
| title_sort | effect of data augmentation on deep learning based segmentation of long axis cine mri |
| topic | cine-MRI long axis ENet data augmentation LVEF heart segmentation |
| url | https://www.mdpi.com/1999-4893/17/1/10 |
| work_keys_str_mv | AT francoislegrand effectofdataaugmentationondeeplearningbasedsegmentationoflongaxiscinemri AT richardmacwan effectofdataaugmentationondeeplearningbasedsegmentationoflongaxiscinemri AT alainlalande effectofdataaugmentationondeeplearningbasedsegmentationoflongaxiscinemri AT lisametairie effectofdataaugmentationondeeplearningbasedsegmentationoflongaxiscinemri AT thomasdecourselle effectofdataaugmentationondeeplearningbasedsegmentationoflongaxiscinemri |