Semantic-Embedding and Shape-Aware U-Net for Ultrasound Eyeball Segmentation

Lin, Fanchao; Liu, Chuanbin; Xie, Hongtao; Zha, Zheng-Jun; Zhang, Yongdong

doi:10.1109/icme.2019.00158

Cited by 7 publications

(2 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the past decades, the deep convolutional neural networks (CNN) [21,22,29] have achieved tremendous success in computer vision tasks. For medical image segmentation, most of CNN-based networks [32] are derived from classical encoder-decoder architectures, known as FCN [34] and UNet [38]. This class of design has been proved to be suitable for dense pixel (voxel)-wise prediction.…”

Section: Segmentation Networkmentioning

confidence: 99%

Weakly-supervised Cerebrovascular Segmentation Network with Shape Prior and Model Indicator

Chen

Huang

et al. 2022

Proceedings of the 2022 International Conference on Multimedia Retrieval

View full text Add to dashboard Cite

Labeling cerebral vessels requires domain knowledge in neurology and could be extremely laborious, and there is a scarcity of public annotated cerebrovascular datasets. Traditional machine learning or statistical models could yield decent results on thick vessels with high contrast while having poor performance on those regions of low contrast. In our work, we employ a statistic model as noisy labels and propose a Transformer-based architecture which utilizes Hessian shape prior as soft supervision. It enhances the learning ability of the network to tubular structures, so that the model can make more accurate predictions on refined cerebrovascular segmentation. Furthermore, to combat the overfitting towards noisy labels as model training, we introduce an effective label extension strategy that only calls for a few manual strokes on one sample. These supplementary labels are not used for supervision but only as an indicator to tell where the model keeps the most generalization capability, so as to further guide the model selection in validation. Our experiments are carried out on a public TOF-MRA dataset from MIDAS data platform, and the results demonstrate that our method shows superior performance on cerebrovascular segmentation which achieves Dice of 0.831±0.040 in the dataset.

show abstract

Section: Segmentation Networkmentioning

confidence: 99%

Weakly-supervised Cerebrovascular Segmentation Network with Shape Prior and Model Indicator

Chen

Huang

et al. 2022

Proceedings of the 2022 International Conference on Multimedia Retrieval

View full text Add to dashboard Cite

show abstract

“…The dense connections contributed to increasing the number of trainable parameters and feature reuse. Lin et al [18] proposed a semantic embedding and shape-aware U-Net model (SSU-Net) for eyeball segmentation, where they used a Signed Distance Field (SDF) instead of a binary mask as the label to learn the shape information and semantic embedding module to combine semantic features at coarser levels. Byra et al [3] modified the U-Net utilizing selective kernels (SKU-Net) for breast mass segmentation in US images by replacing each convolution layer with an SK module with two branches, one of which generates feature maps using dilated convolutions and the other without dilation.…”

Section: Introductionmentioning

confidence: 99%

Slim U-Net: Efficient Anatomical Feature Preserving U-net Architecture for Ultrasound Image Segmentation

Raina

Verma

Chandrashekhara

et al. 2023

Preprint

View full text Add to dashboard Cite

We investigate the applicability of U-Net based models for segmenting Urinary Bladder (UB) in male pelvic view UltraSound (US) images. The segmentation of UB in the US image aids radiologists in diagnosing the UB. However, UB in US images has arbitrary shapes, indistinct boundaries and considerably large inter- and intra-subject variability, making segmentation a quite challenging task. Our study of the state-of-the-art (SOTA) segmentation network, U-Net, for the problem reveals that it often fails to capture the salient characteristics of UB due to the varying shape and scales of anatomy in the noisy US image. Also, U-net has an excessive number of trainable parameters, reporting poor computational efficiency during training. We propose a Slim U-Net to address the challenges of UB segmentation. Slim U-Net proposes to efficiently preserve the salient features of UB by reshaping the structure of U-Net using a less number of 2D convolution layers in the contracting path, in order to preserve and impose them on expanding path. To effectively distinguish the blurred boundaries, we propose a novel annotation methodology, which includes the background area of the image at the boundary of a marked region of interest (RoI), thereby steering the model's attention towards boundaries. In addition, we suggested a combination of loss functions for network training in the complex segmentation of UB. The experimental results demonstrate that Slim U-net is statistically superior to U-net for UB segmentation. The Slim U-net further decreases the number of trainable parameters and training time by 54% and 57.7%, respectively, compared to the standard U-Net, without compromising the segmentation accuracy.

show abstract

CTumorGAN: a unified framework for automatic computed tomography tumor segmentation

Pang

Orgun

et al. 2020

Eur J Nucl Med Mol Imaging

View full text Add to dashboard Cite

Semantic-Embedding and Shape-Aware U-Net for Ultrasound Eyeball Segmentation

Cited by 7 publications

References 6 publications

Weakly-supervised Cerebrovascular Segmentation Network with Shape Prior and Model Indicator

Weakly-supervised Cerebrovascular Segmentation Network with Shape Prior and Model Indicator

Slim U-Net: Efficient Anatomical Feature Preserving U-net Architecture for Ultrasound Image Segmentation

CTumorGAN: a unified framework for automatic computed tomography tumor segmentation

Contact Info

Product

Resources

About