Multicenter UWF-SLO Vessel Segmentation (MU-VS) dataset
收藏资源简介:
本研究创建了首个多中心UWF-SLO血管分割数据集(MU-VS),包含来自两个医院的60张UWF-SLO图像,旨在支持跨中心血管分割研究。数据集涵盖多种疾病类别,由专业眼科医生使用Photoshop软件手动标注血管。创建过程采用了一种高效的Cascade Uncertainty-Predominance (CUP)选择策略,以减少标注负担。该数据集的应用领域主要集中在提高UWF-SLO图像中血管分割的准确性,以辅助眼科疾病的诊断和治疗。
This study developed the first multi-center UWF-SLO vascular segmentation dataset (MU-VS), which includes 60 UWF-SLO images from two hospitals and is intended to support cross-center vascular segmentation research. The dataset covers multiple disease categories, with blood vessels manually annotated by professional ophthalmologists using Photoshop software. An efficient Cascade Uncertainty-Predominance (CUP) selection strategy was adopted during the dataset construction process to reduce the annotation burden. The main application of this dataset focuses on improving the accuracy of vascular segmentation in UWF-SLO images, so as to assist the diagnosis and treatment of ophthalmic diseases.
SFADA-UWF-SLO 数据集概述
简介
本项目引入了一种新的医学图像分割设置,称为源无主动域适应(SFADA)。SFADA旨在促进跨中心的医学图像分割,同时保护数据隐私并减少医疗专业人员的工作量。通过仅需要最少的标注工作,SFADA实现了有效的模型转移,并取得了与完全监督方法相当的结果。
数据集
感谢您的兴趣,代码和数据正在整理中,即将上线!
引用
如果您发现我们的工作对您的研究有用或相关,请考虑引用:
@article{wang2024advancing, title={Advancing UWF-SLO Vessel Segmentation with Source-Free Active Domain Adaptation and a Novel Multi-Center Dataset}, author={Wang, Hongqiu and Luo, Xiangde and Chen, Wu and Tang, Qingqing and Xin, Mei and Wang, Qiong and Zhu, Lei}, journal={arXiv preprint arXiv:2406.13645}, year={2024} }

- 1Advancing UWF-SLO Vessel Segmentation with Source-Free Active Domain Adaptation and a Novel Multi-Center Dataset香港科技大学(广州) · 2024年



