arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2303.09233eess.IVcs.CV
收藏

SwinVFTR:一种用于三维OCT液分割的新型体积特征学习Transformer

SwinVFTR: A Novel Volumetric Feature-learning Transformer for 3D OCT Fluid Segmentation

University of Nevada, Reno(内华达大学雷诺分校);School of Medicine, University of Nevada, Reno(内华达大学雷诺分校医学院)

Khondker Fariha Hossain, Sharif Amit Kamran, Alireza Tavakkoli, George Bebis, Sal Baker

AI总结:

本文提出SwinVFTR,一种基于通道级体积采样、移位窗口Transformer和体积注意力块的三维OCT积液分割架构,在多个OCT数据集上取得了优于现有模型的dice、IOU和SSIM性能。

摘要:

在三维光学相干断层扫描(OCT)图像中准确分割积液对于检测眼部疾病至关重要,但仍然具有挑战性。传统的基于自编码器的方法难以应对分辨率损失和信息恢复问题。虽然基于Transformer的模型改善了分割性能,但它们并未针对三维OCT体积进行优化,而三维OCT体积因供应商和提取技术而异。为了解决这一问题,我们提出了SwinVFTR,这是一种用于三维OCT图像中精确积液分割的Transformer架构。SwinVFTR采用通道级体积采样和移位窗口Transformer块来改善积液定位。此外,一种新颖的体积注意力块增强了空间和深度方向的注意力。使用多类dice损失进行训练后,SwinVFTR在Spectralis、Cirrus和Topcon OCT数据集上优于现有模型,分别实现了0.72、0.59和0.68的平均dice分数,并在平均交并比(IOU)和结构相似性(SSIM)指标上取得了优越性能。

英文摘要:

Accurately segmenting fluid in 3D optical coherence tomography (OCT) images is critical for detecting eye diseases but remains challenging. Traditional autoencoder-based methods struggle with resolution loss and information recovery. While transformer-based models improve segmentation, they arent optimized for 3D OCT volumes, which vary by vendor and extraction technique. To address this, we propose SwinVFTR, a transformer architecture for precise fluid segmentation in 3D OCT images. SwinVFTR employs channel-wise volumetric sampling and a shifted window transformer block to improve fluid localization. Moreover, a novel volumetric attention block enhances spatial and depth-wise attention. Trained using multi-class dice loss, SwinVFTR outperforms existing models on Spectralis, Cirrus, and Topcon OCT datasets, achieving mean dice scores of 0.72, 0.59, and 0.68, respectively, along with superior performance in mean intersection-over-union (IOU) and structural similarity (SSIM) metrics.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑