CS-Net: Channel and Spatial Attention Network for Curvilinear Structure Segmentation

Mou, Lei; Zhao, Yitian; Chen, Li; Cheng, Jun; Gu, Zaiwang; Hao, Honggang; Qi, Hong; Zheng, Yalin; Frangi, Alejandro F.; Liu, Jiang

doi:10.1007/978-3-030-32239-7_80

Cited by 162 publications

(137 citation statements)

References 18 publications

Supporting

Mentioning

137

Contrasting

Order By: Relevance

“…As we can see from Table 1, the Self-attention model [8,10] drops two percentage points over multi-kernel context encoder networks [13]. Very recently, both DANet [17] and CS-Net [18] have exploited two types of self-attention models acting on the top of a feature encoder path from different pretrained network architectures with better performance than Attention UNet [9] and CE-Net [13], for example, 79.97% vs 78.90% in Dice, 30.94 pixels vs 32.90 pixels in HD, 74.50% vs 73.03% in Precision and 90.38% vs 90.03% in Recall. Moreover, these good segmentation results also illustrate that diverse self-attention strategies can further boost the feature representative capability of a model for accurate tumor localization and segmentation.…”

Section: Quantitative Analysismentioning

confidence: 97%

“…(2) Context Based Model: R2U-Net [11] utilizes recurrent and residual networks; CE-Net [13] embeds a multi-kernel context encoding mechanism like Inception architecture; Self-attention [8,10] exploits spatial context information. And (3) Attention Based Model: SENet [15] uses channel attention mechanism; both DANet [17] and CS-Net [18] place self-attention schemes on the top of encoder stage, but with different network architectures. (4) Fused Model: Attention UNet [9] and Self-attention [8,10].…”

Section: Data and Implementation Detailsmentioning

confidence: 99%

See 1 more Smart Citation

Correlation Matters: Multi-scale Fine-Grained Contextual Information Extraction for Hepatic Tumor Segmentation

Pang

et al. 2020

Advances in Knowledge Discovery and Data Mining

View full text Add to dashboard Cite

Automatic tumor segmentation has been used as a diagnostic aid in the identification of diseases such as tumors from liver CT scans, and their treatment. Owing to their success in computer vision tasks, the state-of-the-art Fully Convolutional Networks (FCNs) or U-Net based models have often been employed in many recent studies for automatic tumor segmentation to learn numerous weight-shared convolutional kernels and extract various semantic features. However, the correlation between different tumor regions in feature maps cannot be easily captured due to the lack of contextual dependencies, which in turn limits the representative capability of the adopted models and thus affects the accuracy of tumor segmentation results. To resolve this issue, we propose a novel framework for segmentation of tumors in liver CT scans, which can explicitly extract multi-scale fine-grained contextual information by adaptively aggregating local features with their global dependencies. The proposed multi-scale framework features a light model with a very few additional parameters, and also its visualization capability significantly boosts networks' interpretability. Experimental results on a real-world liver tumor CT dataset illustrate that the proposed framework achieves the state-of-the-art performance in terms of a number of widely used evaluation criteria for the hepatic tumor segmentation task.

show abstract

Section: Quantitative Analysismentioning

confidence: 97%

Section: Data and Implementation Detailsmentioning

confidence: 99%

Correlation Matters: Multi-scale Fine-Grained Contextual Information Extraction for Hepatic Tumor Segmentation

Pang

et al. 2020

Advances in Knowledge Discovery and Data Mining

View full text Add to dashboard Cite

show abstract

“…Deep learning methods have demonstrated superior performance and better prospects for many medical image segmentation problems. In this work, we employed one of the most latest curvilinear structure segmentation network: CS-Net [55], for fully automatic segmentation of corneal nerves, with and without application of image enhancement methods. We trained the CS-Net on randomly sampled 80% images from CCM-A, leaving out 20% of this dataset as a testing set.…”

Section: ) Image Enhancement-guided Fiber Segmentationmentioning

confidence: 99%

Automated Tortuosity Analysis of Nerve Fibers in Corneal Confocal Microscopy

Zhao

Zhang

Pereira

et al. 2020

IEEE Trans. Med. Imaging

Self Cite

View full text Add to dashboard Cite

Precise characterization and analysis of corneal nerve fiber tortuosity are of great importance in facilitating examination and diagnosis of many eye-related diseases. In this paper we propose a fully automated method for image-level tortuosity estimation, comprising image enhancement, exponential curvature estimation, and tortuosity level classification. The image enhancement component is based on an extended Retinex model, which not only corrects imbalanced illumination and improves image contrast in an image, but also models noise explicitly to aid removal of imaging noise. Afterwards, we take advantage of exponential curvature estimation in the 3D space of positions and orientations to directly measure curvature based on the enhanced images, rather than relying on the explicit segmentation and skeletonization steps in a conventional pipeline usually with accumulated preprocessing errors. The proposed method has been applied over two corneal nerve microscopy datasets for the estimation of a tortuosity level for each image. The experimental results show that it performs better than several selected state-of-the-art methods. Furthermore, we have performed manual gradings at tortuosity level of four hundred and three corneal nerve microscopic images, and this dataset has been released for public access to facilitate other researchers in the community in carrying out further research on the same and related topics.

show abstract

“…Guo et al [ 20 ] used the residual block in the channel attention mechanism and proposed that the channel attention residual block improves the recognition ability of the network. Mou et al [ 21 ] used a self-attention mechanism in the encoder and decoder to combine local features and global correlation. However, these attention mechanisms do not take the impact of multi-scale image features on the attention gate into account, and the channel dependence between different scales is ignored.…”

Section: Introductionmentioning

confidence: 99%

“…Inspired by the successful application of the channel attention mechanism in the field of medical image segmentation [ 19 , 20 , 21 ], we introduced an aggregation channel attention network to improve the performance of optic disc segmentation of fundus images. First, in order to alleviate the disappearance of gradients and reduce the number of parameters [ 22 ], we use DenseNet blocks to extract high-level features.…”

Section: Introductionmentioning

confidence: 99%

Optic Disc Segmentation Using Attention-Based U-Net and the Improved Cross-Entropy Convolutional Neural Network

Jin

Wang

et al. 2020

Entropy

View full text Add to dashboard Cite

Medical image segmentation is an important part of medical image analysis. With the rapid development of convolutional neural networks in image processing, deep learning methods have achieved great success in the field of medical image processing. Deep learning is also used in the field of auxiliary diagnosis of glaucoma, and the effective segmentation of the optic disc area plays an important assistant role in the diagnosis of doctors in the clinical diagnosis of glaucoma. Previously, many U-Net-based optic disc segmentation methods have been proposed. However, the channel dependence of different levels of features is ignored. The performance of fundus image segmentation in small areas is not satisfactory. In this paper, we propose a new aggregation channel attention network to make full use of the influence of context information on semantic segmentation. Different from the existing attention mechanism, we exploit channel dependencies and integrate information of different scales into the attention mechanism. At the same time, we improved the basic classification framework based on cross entropy, combined the dice coefficient and cross entropy, and balanced the contribution of dice coefficients and cross entropy loss to the segmentation task, which enhanced the performance of the network in small area segmentation. The network retains more image features, restores the significant features more accurately, and further improves the segmentation performance of medical images. We apply it to the fundus optic disc segmentation task. We demonstrate the segmentation performance of the model on the Messidor dataset and the RIM-ONE dataset, and evaluate the proposed architecture. Experimental results show that our network architecture improves the prediction performance of the base architectures under different datasets while maintaining the computational efficiency. The results render that the proposed technologies improve the segmentation with 0.0469 overlapping error on Messidor.

show abstract

CS-Net: Channel and Spatial Attention Network for Curvilinear Structure Segmentation

Cited by 162 publications

References 18 publications

Correlation Matters: Multi-scale Fine-Grained Contextual Information Extraction for Hepatic Tumor Segmentation

Correlation Matters: Multi-scale Fine-Grained Contextual Information Extraction for Hepatic Tumor Segmentation

Automated Tortuosity Analysis of Nerve Fibers in Corneal Confocal Microscopy

Optic Disc Segmentation Using Attention-Based U-Net and the Improved Cross-Entropy Convolutional Neural Network

Contact Info

Product

Resources

About