Lightweight image segmentation for echocardiography
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911137091551232 |
|---|---|
| author | Kjelsrud, Anders Løvstakken, Lasse Smistad, Erik Dalen, Håvard Van De Vyver, Gilles |
| author_facet | Kjelsrud, Anders Løvstakken, Lasse Smistad, Erik Dalen, Håvard Van De Vyver, Gilles |
| contents | Accurate segmentation of the left ventricle in echocardiography can enable fully automatic extraction of clinical measurements such as volumes and ejection fraction. While models configured by nnU-Net perform well, they are large and slow, thus limiting real-time use. We identified the most effective components of nnU-Net for cardiac segmentation through an ablation study, incrementally evaluating data augmentation schemes, architectural modifications, loss functions, and post-processing techniques. Our analysis revealed that simple affine augmentations and deep supervision drive performance, while complex augmentations and large model capacity offer diminishing returns. Based on these insights, we developed a lightweight U-Net (2M vs 33M parameters) that achieves statistically equivalent performance to nnU-Net on CAMUS (N=500) with Dice scores of 0.93/0.85/0.89 vs 0.93/0.86/0.89 for LV/MYO/LA ($p>0.05$), while being 16 times smaller and 4 times faster (1.35ms vs 5.40ms per frame) than the default nnU-Net configuration. Cross-dataset evaluation on an internal dataset (N=311) confirms comparable generalization. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_03631 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Lightweight image segmentation for echocardiography Kjelsrud, Anders Løvstakken, Lasse Smistad, Erik Dalen, Håvard Van De Vyver, Gilles Computer Vision and Pattern Recognition Accurate segmentation of the left ventricle in echocardiography can enable fully automatic extraction of clinical measurements such as volumes and ejection fraction. While models configured by nnU-Net perform well, they are large and slow, thus limiting real-time use. We identified the most effective components of nnU-Net for cardiac segmentation through an ablation study, incrementally evaluating data augmentation schemes, architectural modifications, loss functions, and post-processing techniques. Our analysis revealed that simple affine augmentations and deep supervision drive performance, while complex augmentations and large model capacity offer diminishing returns. Based on these insights, we developed a lightweight U-Net (2M vs 33M parameters) that achieves statistically equivalent performance to nnU-Net on CAMUS (N=500) with Dice scores of 0.93/0.85/0.89 vs 0.93/0.86/0.89 for LV/MYO/LA ($p>0.05$), while being 16 times smaller and 4 times faster (1.35ms vs 5.40ms per frame) than the default nnU-Net configuration. Cross-dataset evaluation on an internal dataset (N=311) confirms comparable generalization. |
| title | Lightweight image segmentation for echocardiography |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2509.03631 |