Channel Attention Residual U-Net for Retinal Vessel Segmentation
- URL: http://arxiv.org/abs/2004.03702v5
- Date: Tue, 20 Oct 2020 19:48:55 GMT
- Title: Channel Attention Residual U-Net for Retinal Vessel Segmentation
- Authors: Changlu Guo, M\'arton Szemenyei, Yangtao Hu, Wenle Wang, Wei Zhou,
Yugen Yi
- Abstract summary: We propose a new deep learning model, namely Channel Attention Residual U-Net (CAR-UNet) to accurately segment retinal vascular and non-vascular pixels.
The results show that our proposed CAR-UNet has reached the state-of-the-art performance on three publicly available retinal vessel datasets.
- Score: 8.109768170171357
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: Retinal vessel segmentation is a vital step for the diagnosis of many early
eye-related diseases. In this work, we propose a new deep learning model,
namely Channel Attention Residual U-Net (CAR-UNet), to accurately segment
retinal vascular and non-vascular pixels. In this model, we introduced a novel
Modified Efficient Channel Attention (MECA) to enhance the discriminative
ability of the network by considering the interdependence between feature maps.
On the one hand, we apply MECA to the "skip connections" in the traditional
U-shaped networks, instead of simply copying the feature maps of the
contracting path to the corresponding expansive path. On the other hand, we
propose a Channel Attention Double Residual Block (CADRB), which integrates
MECA into a residual structure as a core structure to construct the proposed
CAR-UNet. The results show that our proposed CAR-UNet has reached the
state-of-the-art performance on three publicly available retinal vessel
datasets: DRIVE, CHASE DB1 and STARE.
Related papers
- CB-HVTNet: A channel-boosted hybrid vision transformer network for
lymphocyte assessment in histopathological images [0.40777876591043144]
We propose a Channel Boosted Hybrid Vision Transformer (CB HVT) that uses transfer learning to generate boosted channels and employs both transformers and CNNs to analyse lymphocytes in medical images.
CB HVT comprises five modules, including a channel generation module, channel exploitation module, channel merging module, region-aware module, and a detection and segmentation head.
The results show that CB HVT outperformed other state of the art detection models, and has good generalization ability, demonstrating its value as a tool for pathologists.
arXiv Detail & Related papers (2023-05-16T06:40:04Z) - Graph Convolution Based Cross-Network Multi-Scale Feature Fusion for
Deep Vessel Segmentation [36.12106059822428]
Vessels reconstructed using existing methods are often not sufficiently accurate to meet clinical use standards.
Our network consists of two cascadedworks performing initial and refined segmentation respectively.
Cross-network multi-scale feature fusion is performed between these two U-shaped networks to effectively support high-quality vessel segmentation.
arXiv Detail & Related papers (2023-01-06T05:56:50Z) - Dual-Stage Deeply Supervised Attention-based Convolutional Neural
Networks for Mandibular Canal Segmentation in CBCT Scans [4.140750794848906]
We propose a novel dual-stage deep learning based scheme for automatic detection of mandibular canal.
Particularly, we first we enhance the CBCT scans by employing the novel histogram-based dynamic windowing scheme.
After enhancement, we design 3D deeply supervised attention U-Net architecture for localize the volume of interest.
arXiv Detail & Related papers (2022-10-06T09:08:56Z) - Channelized Axial Attention for Semantic Segmentation [70.14921019774793]
We propose the Channelized Axial Attention (CAA) to seamlessly integratechannel attention and axial attention with reduced computationalcomplexity.
Our CAA not onlyrequires much less computation resources compared with otherdual attention models such as DANet, but also outperforms the state-of-the-art ResNet-101-based segmentation models on alltested datasets.
arXiv Detail & Related papers (2021-01-19T03:08:03Z) - Channel-wise Knowledge Distillation for Dense Prediction [73.99057249472735]
We propose to align features channel-wise between the student and teacher networks.
We consistently achieve superior performance on three benchmarks with various network structures.
arXiv Detail & Related papers (2020-11-26T12:00:38Z) - Convolutional Neural Network optimization via Channel Reassessment
Attention module [19.566271646280978]
We propose a novel network optimization module called Channel Reassessment (CRA) module.
CRA module uses channel attentions with spatial information of feature maps to enhance representational power of networks.
Experiments on ImageNet and MS datasets demonstrate that embedding CRA module on various networks effectively improves the performance under different evaluation standards.
arXiv Detail & Related papers (2020-10-12T11:27:17Z) - Rethinking the Extraction and Interaction of Multi-Scale Features for
Vessel Segmentation [53.187152856583396]
We propose a novel deep learning model called PC-Net to segment retinal vessels and major arteries in 2D fundus image and 3D computed tomography angiography (CTA) scans.
In PC-Net, the pyramid squeeze-and-excitation (PSE) module introduces spatial information to each convolutional block, boosting its ability to extract more effective multi-scale features.
arXiv Detail & Related papers (2020-10-09T08:22:54Z) - Adaptive feature recombination and recalibration for semantic
segmentation with Fully Convolutional Networks [57.64866581615309]
We propose recombination of features and a spatially adaptive recalibration block that is adapted for semantic segmentation with Fully Convolutional Networks.
Results indicate that Recombination and Recalibration improve the results of a competitive baseline, and generalize across three different problems.
arXiv Detail & Related papers (2020-06-19T15:45:03Z) - Ventral-Dorsal Neural Networks: Object Detection via Selective Attention [51.79577908317031]
We propose a new framework called Ventral-Dorsal Networks (VDNets)
Inspired by the structure of the human visual system, we propose the integration of a "Ventral Network" and a "Dorsal Network"
Our experimental results reveal that the proposed method outperforms state-of-the-art object detection approaches.
arXiv Detail & Related papers (2020-05-15T23:57:36Z) - Multi-level Context Gating of Embedded Collective Knowledge for Medical
Image Segmentation [32.96604621259756]
We propose an extension of U-Net for medical image segmentation.
We take full advantages of U-Net, Squeeze and Excitation (SE) block, bi-directional ConvLSTM (BConvLSTM), and the mechanism of dense convolutions.
The proposed model is evaluated on six datasets.
arXiv Detail & Related papers (2020-03-10T12:29:59Z) - Channel Equilibrium Networks for Learning Deep Representation [63.76618960820138]
This work shows that the combination of normalization and rectified linear function leads to inhibited channels.
Unlike prior arts that simply removed the inhibited channels, we propose to "wake them up" during training by designing a novel neural building block.
Channel Equilibrium (CE) block enables channels at the same layer to contribute equally to the learned representation.
arXiv Detail & Related papers (2020-02-29T09:02:31Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.