Data augmentation via warping transforms for modeling natural variability in the corneal endothelium enhances semi-supervised segmentation.
Image segmentation of the corneal endothelium with deep convolutional neural networks (CNN) is challenging due to the scarcity of expert-annotated data. This work proposes a data augmentation technique via warping to enhance the performance of semi-supervised training of CNNs for accurate segmentati...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Public Library of Science (PLoS)
2024-01-01
|
| Series: | PLoS ONE |
| Online Access: | https://doi.org/10.1371/journal.pone.0311849 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1846163598740553728 |
|---|---|
| author | Sergio Sanchez Noelia Vallez Gloria Bueno Andres G Marrugo |
| author_facet | Sergio Sanchez Noelia Vallez Gloria Bueno Andres G Marrugo |
| author_sort | Sergio Sanchez |
| collection | DOAJ |
| description | Image segmentation of the corneal endothelium with deep convolutional neural networks (CNN) is challenging due to the scarcity of expert-annotated data. This work proposes a data augmentation technique via warping to enhance the performance of semi-supervised training of CNNs for accurate segmentation. We use a unique augmentation process for images and masks involving keypoint extraction, Delaunay triangulation, local affine transformations, and mask refinement. This approach accurately captures the natural variability of the corneal endothelium, enriching the dataset with realistic and diverse images. The proposed method achieved an increase in the mean intersection over union (mIoU) and Dice coefficient (DC) metrics of 17.2% and 4.8% respectively, for the segmentation task in corneal endothelial images on multiple CNN architectures. Our data augmentation strategy successfully models the natural variability in corneal endothelial images, thereby enhancing the performance and generalization capabilities of semi-supervised CNNs in medical image cell segmentation tasks. |
| format | Article |
| id | doaj-art-1e528f7041374a4da7dc61963eb675f2 |
| institution | Kabale University |
| issn | 1932-6203 |
| language | English |
| publishDate | 2024-01-01 |
| publisher | Public Library of Science (PLoS) |
| record_format | Article |
| series | PLoS ONE |
| spelling | doaj-art-1e528f7041374a4da7dc61963eb675f22024-11-19T05:31:44ZengPublic Library of Science (PLoS)PLoS ONE1932-62032024-01-011911e031184910.1371/journal.pone.0311849Data augmentation via warping transforms for modeling natural variability in the corneal endothelium enhances semi-supervised segmentation.Sergio SanchezNoelia VallezGloria BuenoAndres G MarrugoImage segmentation of the corneal endothelium with deep convolutional neural networks (CNN) is challenging due to the scarcity of expert-annotated data. This work proposes a data augmentation technique via warping to enhance the performance of semi-supervised training of CNNs for accurate segmentation. We use a unique augmentation process for images and masks involving keypoint extraction, Delaunay triangulation, local affine transformations, and mask refinement. This approach accurately captures the natural variability of the corneal endothelium, enriching the dataset with realistic and diverse images. The proposed method achieved an increase in the mean intersection over union (mIoU) and Dice coefficient (DC) metrics of 17.2% and 4.8% respectively, for the segmentation task in corneal endothelial images on multiple CNN architectures. Our data augmentation strategy successfully models the natural variability in corneal endothelial images, thereby enhancing the performance and generalization capabilities of semi-supervised CNNs in medical image cell segmentation tasks.https://doi.org/10.1371/journal.pone.0311849 |
| spellingShingle | Sergio Sanchez Noelia Vallez Gloria Bueno Andres G Marrugo Data augmentation via warping transforms for modeling natural variability in the corneal endothelium enhances semi-supervised segmentation. PLoS ONE |
| title | Data augmentation via warping transforms for modeling natural variability in the corneal endothelium enhances semi-supervised segmentation. |
| title_full | Data augmentation via warping transforms for modeling natural variability in the corneal endothelium enhances semi-supervised segmentation. |
| title_fullStr | Data augmentation via warping transforms for modeling natural variability in the corneal endothelium enhances semi-supervised segmentation. |
| title_full_unstemmed | Data augmentation via warping transforms for modeling natural variability in the corneal endothelium enhances semi-supervised segmentation. |
| title_short | Data augmentation via warping transforms for modeling natural variability in the corneal endothelium enhances semi-supervised segmentation. |
| title_sort | data augmentation via warping transforms for modeling natural variability in the corneal endothelium enhances semi supervised segmentation |
| url | https://doi.org/10.1371/journal.pone.0311849 |
| work_keys_str_mv | AT sergiosanchez dataaugmentationviawarpingtransformsformodelingnaturalvariabilityinthecornealendotheliumenhancessemisupervisedsegmentation AT noeliavallez dataaugmentationviawarpingtransformsformodelingnaturalvariabilityinthecornealendotheliumenhancessemisupervisedsegmentation AT gloriabueno dataaugmentationviawarpingtransformsformodelingnaturalvariabilityinthecornealendotheliumenhancessemisupervisedsegmentation AT andresgmarrugo dataaugmentationviawarpingtransformsformodelingnaturalvariabilityinthecornealendotheliumenhancessemisupervisedsegmentation |