From Deterministic to Probabilistic: A Novel Perspective on Domain Generalization for Medical Image Segmentation
Fuente:
arXiv
Guardado en:
| Autores principales: | , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866909419252482048 |
|---|---|
| author | Xu, Yuheng Zhang, Taiping |
| author_facet | Xu, Yuheng Zhang, Taiping |
| contents | Traditional domain generalization methods often rely on domain alignment to reduce inter-domain distribution differences and learn domain-invariant representations. However, domain shifts are inherently difficult to eliminate, which limits model generalization. To address this, we propose an innovative framework that enhances data representation quality through probabilistic modeling and contrastive learning, reducing dependence on domain alignment and improving robustness under domain variations. Specifically, we combine deterministic features with uncertainty modeling to capture comprehensive feature distributions. Contrastive learning enforces distribution-level alignment by aligning the mean and covariance of feature distributions, enabling the model to dynamically adapt to domain variations and mitigate distribution shifts. Additionally, we design a frequency-domain-based structural enhancement strategy using discrete wavelet transforms to preserve critical structural details and reduce visual distortions caused by style variations. Experimental results demonstrate that the proposed framework significantly improves segmentation performance, providing a robust solution to domain generalization challenges in medical image segmentation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2412_05572 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | From Deterministic to Probabilistic: A Novel Perspective on Domain Generalization for Medical Image Segmentation Xu, Yuheng Zhang, Taiping Computer Vision and Pattern Recognition I.4 Traditional domain generalization methods often rely on domain alignment to reduce inter-domain distribution differences and learn domain-invariant representations. However, domain shifts are inherently difficult to eliminate, which limits model generalization. To address this, we propose an innovative framework that enhances data representation quality through probabilistic modeling and contrastive learning, reducing dependence on domain alignment and improving robustness under domain variations. Specifically, we combine deterministic features with uncertainty modeling to capture comprehensive feature distributions. Contrastive learning enforces distribution-level alignment by aligning the mean and covariance of feature distributions, enabling the model to dynamically adapt to domain variations and mitigate distribution shifts. Additionally, we design a frequency-domain-based structural enhancement strategy using discrete wavelet transforms to preserve critical structural details and reduce visual distortions caused by style variations. Experimental results demonstrate that the proposed framework significantly improves segmentation performance, providing a robust solution to domain generalization challenges in medical image segmentation. |
| title | From Deterministic to Probabilistic: A Novel Perspective on Domain Generalization for Medical Image Segmentation |
| topic | Computer Vision and Pattern Recognition I.4 |
| url | https://arxiv.org/abs/2412.05572 |