Unsupervised Domain Adaptation for Brain Vessel Segmentation through Transwarp Contrastive Learning
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866909118259789824 |
|---|---|
| author | Lin, Fengming Xia, Yan MacRaild, Michael Deo, Yash Dou, Haoran Liu, Qiongyao Wu, Kun Ravikumar, Nishant Frangi, Alejandro F. |
| author_facet | Lin, Fengming Xia, Yan MacRaild, Michael Deo, Yash Dou, Haoran Liu, Qiongyao Wu, Kun Ravikumar, Nishant Frangi, Alejandro F. |
| contents | Unsupervised domain adaptation (UDA) aims to align the labelled source distribution with the unlabelled target distribution to obtain domain-invariant predictive models. Since cross-modality medical data exhibit significant intra and inter-domain shifts and most are unlabelled, UDA is more important while challenging in medical image analysis. This paper proposes a simple yet potent contrastive learning framework for UDA to narrow the inter-domain gap between labelled source and unlabelled target distribution. Our method is validated on cerebral vessel datasets. Experimental results show that our approach can learn latent features from labelled 3DRA modality data and improve vessel segmentation performance in unlabelled MRA modality data. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_15237 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Unsupervised Domain Adaptation for Brain Vessel Segmentation through Transwarp Contrastive Learning Lin, Fengming Xia, Yan MacRaild, Michael Deo, Yash Dou, Haoran Liu, Qiongyao Wu, Kun Ravikumar, Nishant Frangi, Alejandro F. Computer Vision and Pattern Recognition Machine Learning Unsupervised domain adaptation (UDA) aims to align the labelled source distribution with the unlabelled target distribution to obtain domain-invariant predictive models. Since cross-modality medical data exhibit significant intra and inter-domain shifts and most are unlabelled, UDA is more important while challenging in medical image analysis. This paper proposes a simple yet potent contrastive learning framework for UDA to narrow the inter-domain gap between labelled source and unlabelled target distribution. Our method is validated on cerebral vessel datasets. Experimental results show that our approach can learn latent features from labelled 3DRA modality data and improve vessel segmentation performance in unlabelled MRA modality data. |
| title | Unsupervised Domain Adaptation for Brain Vessel Segmentation through Transwarp Contrastive Learning |
| topic | Computer Vision and Pattern Recognition Machine Learning |
| url | https://arxiv.org/abs/2402.15237 |