Channel Attention Residual U-Net for Retinal Vessel Segmentation

  title={Channel Attention Residual U-Net for Retinal Vessel Segmentation},
  author={Changlu Guo and Marton Szemenyei and Yugen Yi and Wei Zhou},
  journal={ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)},
  • Changlu Guo, Marton Szemenyei, W. Zhou
  • Published 7 April 2020
  • Computer Science
  • ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
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… 

Figures and Tables from this paper

DR-VNet: Retinal Vessel Segmentation via Dense Residual UNet

A new deep learning pipeline combining the efficiency of residual dense net blocks and, residual squeeze and excitation blocks is proposed, which outperforms current state of the art techniques on the sensitivity metric relevant to assess capture of small vessels.

Novel Application of Long Short-Term Memory Network for 3D to 2D Retinal Vessel Segmentation in Adaptive Optics—Optical Coherence Tomography Volumes

This study shows that a bidirectional convolutional LSTM module improves standard automated vessel segmentation in AO-OCT volumes, although with higher time cost.

Residual Channel Attention Network for Brain Glioma Segmentation

A novel deep neural network is implemented that integrates residual channel attention modules to calibrate intermediate features for glioma segmentation and adaptively weights feature channel-wise to optimize the latent representation of gliomas.

X-ray coronary centerline extraction based on C-UNet and a multifactor reconnection algorithm

Combining Multiple Style Transfer Networks and Transfer Learning For LGE-CMR Segmentation

An algorithm for segmenting late gadolinium enhancement cardiac magnetic resonance (LGE-CMR) in the absence of labeled training data is presented and the model is fine-tuned with a few LGE images and labels.

MR‐UNet: An UNet model using multi‐scale and residual convolutions for retinal vessel segmentation

An improved MR‐UNet is proposed, which designs two new blocks: the multi‐scale convolution (Multiconv) block and the residual Convolution (Resconv) block, which improves the model's ability to segment small blood vessels.

U-Net: A Smart Application with Multidimensional Attention Network for Remote Sensing Images

This work proposes a new deep learning model, namely Multidimension Attention U-Net (MDAU-Net), to accurately segment building pixels and nonbuilding pixels in remote sensing images and introduces a novel MultidIMension Modified Efficient Channel Attention (MD-MECA) model to enhance the network discriminative ability through considering the interdependence between feature maps.

Nested Multiple Instance Learning with Attention Mechanisms

A Nested Multiple Instance with Attention (NMIA) model architecture is proposed combining the concept of nesting with attention mechanisms and it is shown that NMIA performs as conventional MIL in simple scenarios and can grasp a complex scenario providing insights to the latent labels at different levels.



RCA-U-Net: Residual Channel Attention U-Net for Fast Tissue Quantification in Magnetic Resonance Fingerprinting

A novel deep learning approach, namely residual channel attention U-Net (RCA-U-Net), to perform the tissue quantification task in MRF, which improves the accuracy of T2 quantification with MRF under high acceleration rates as compared to the state-of-the-art methods.

Ridge-based vessel segmentation in color images of the retina

A method is presented for automated segmentation of vessels in two-dimensional color images of the retina based on extraction of image ridges, which coincide approximately with vessel centerlines, which is compared with two recently published rule-based methods.

An Ensemble Classification-Based Approach Applied to Retinal Blood Vessel Segmentation

This method uses an ensemble system of bagged and boosted decision trees and utilizes a feature vector based on the orientation analysis of gradient vector field, morphological transformation, line strength measures, and Gabor filter responses to segmentation of blood vessels in retinal photographs.

DropBlock: A regularization method for convolutional networks

DropBlock is introduced, a form of structured dropout, where units in a contiguous region of a feature map are dropped together, and it is found that applying DropbBlock in skip connections in addition to the convolution layers increases the accuracy.

Deep Residual Learning for Image Recognition

This work presents a residual learning framework to ease the training of networks that are substantially deeper than those used previously, and provides comprehensive empirical evidence showing that these residual networks are easier to optimize, and can gain accuracy from considerably increased depth.

Dense Residual Network for Retinal Vessel Segmentation

This work proposes an efficient method to segment blood vessels in Scanning Laser Ophthalmoscopy (SLO) retinal images using a deep dense residual network structure called DRNet, inspired by U-Net, "feature map reuse" and residual learning, and introduces DropBlock to alleviate the overfitting problem of the network.

CABNet: Category Attention Block for Imbalanced Diabetic Retinopathy Grading

A novel Category Attention Block (CAB), which explores more discriminative region-wise features for each DR grade and treats each category equally, and the GAB, which can exploit detailed and class-agnostic global attention feature maps for fundus images.

SD-Unet: A Structured Dropout U-Net for Retinal Vessel Segmentation

This paper utilizes the U-shaped structure to exploit the local features of the retinal vessels and perform retinal vessel segmentation in an end-to-end manner and proposes a new method called Structured Dropout U-Net (SD-Unet), which abandons the traditional dropout for convolutional layers, and applies the structured dropout to regularize U- net.

Dense Dilated Network With Probability Regularized Walk for Vessel Detection

The proposed novel method for retinal vessel detection includes a dense dilated network to get an initial detection of the vessels and a probability regularized walk algorithm to address the fracture issue in the initial detection.

Dual Encoding U-Net for Retinal Vessel Segmentation

A novel Dual Encoding U-Net (DEU-Net), which has two encoders: a spatial path with large kernel to preserve the spatial information and a context path with multiscale convolution block to capture more semantic information, and a feature fusion module to combine the different level of feature representation.