MAU-Net: A Retinal Vessels Segmentation Method

Li, Han; Wang, Yikuang; Wan, Cheng; Shen, Jianxin; Chen, Zhiqiang; Ye, Hui; Yu, Qingliang

doi:10.1109/embc44109.2020.9176093

Cited by 15 publications

(8 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Within this context, various well-known networks have been extended in order to enhance the segmentation results, as the case of U-net architecture. Certain extension consists of applying advanced convolution layers, as the case of [10,11] and [12], where standard ones have been replaced respectively by dilated convolution layers and deformable convolution layer. Their aim is respectively to enlarge the convolution receptive field for learning more distributed information and to adapting the convolution receptive field form to the vessel structure.…”

Section: Related Workmentioning

confidence: 99%

Extended U-net for Retinal Vessel Segmentation

Boudegga

Elloumi

Kachouri

et al. 2022

Advances in Computational Collective Intelligence

View full text Add to dashboard Cite

The retinal vascular tree is an important biomarker for the diagnosis of ocular dis-ease, where an efficient segmentation is highly required. Recently, various standard Convolutional Neural Networks CNN dedicated for segmentation are applied for retinal vessel segmentation. In fact, retinal blood vessels are presented in different retinal image resolutions with a complicated morphology. Thus, it is difficult for the standard configuration of CNN to guarantee an optimal feature extraction and efficient segmentation whatever the image resolution is. In this paper, new retinal vessel segmentation approach based on deep learning architecture is propounded. The idea consists of enlarging the kernel size of convolution layer in order to cover the vessel pixels as well as more neighbors for extracting features. Within this objective, our main contribution consists of identifying the kernel size in correlation with retinal image resolution through an experimental approach. Then, a novel U-net extension is proposed by using convolution layer with the identified kernel size. The suggested method is evaluated on two public databases DRIVE and HRF having different resolutions, where higher segmentation performances are achieved respectively with 5*5 and 7*7 convolution kernel sizes. The average accuracy and sensitivity values for DRIVE and HRF databases are respectively in the order of to 0.9785, 0.8474 and 0.964 and 0.803 which outperform the segmentation performance for the standard U-net.

show abstract

Section: Related Workmentioning

confidence: 99%

Extended U-net for Retinal Vessel Segmentation

Boudegga

Elloumi

Kachouri

et al. 2022

Advances in Computational Collective Intelligence

View full text Add to dashboard Cite

show abstract

“…Where L n is the number of all scales, and igc I is the pixel value of the corresponding pixel of the original green channel image. Then use equation (2) to fuse the response functions at different scales to get the result. This method solves the dilation phenomenon that occurs when a blood vessel merges into a single blood vessel and intersects the blood vessels.…”

Section: Analysis Of the Original Color Retinal Imagementioning

confidence: 99%

“…By detecting changes in the structure of blood vessel width, angle, and branches, it can help diagnose diseases such as diabetes, glaucoma, and hypertension. [1][2][3] The retinal vascular network is a tree-like structure with many branches, and the small blood vessels in the branches have a small contrast with the background, and the contour boundaries are blurred, which makes the automatic segmentation of small blood vessels more difficult. [4][5][6] Therefore, this paper proposes a retinal vessel segmentation method based on a multi-scale linear detector combining local enhancement and global enhancement.…”

Section: Introductionmentioning

confidence: 99%

Construction and application of color fundus image segmentation algorithm based on Multi-Scale local combined global enhancement

2021

View full text Add to dashboard Cite

Objective: Aiming at the problem of low accuracy in extracting small blood vessels from existing retinal blood vessel images, a retinal blood vessel segmentation method based on a combination of a multi-scale linear detector and local and global enhancement is proposed. Methods: The multi-scale line detector is studied, and it is divided into two parts: small scale and large scale. The small scale is used to detect the locally enhanced image and the large scale is used to detect the globally enhanced image. Fusion the response functions at different scales to get the final retinal vascular structure. Results: Experiments on two databases STARE and DRIVE, show that the average vascular accuracy rates obtained by the algorithm reach 96.62% and 96.45%, and the average true positive rates reach 75.52% and 83.07%, respectively. Conclusion: The segmentation accuracy is high, and better blood vessel segmentation results can be obtained. doi: https://doi.org/10.12669/pjms.37.6-WIT.4848 How to cite this:Hao Y, Xie H, Qiu R. Construction and application of color fundus image segmentation algorithm based on Multi-Scale local combined global enhancement. Pak J Med Sci. 2021;37(6):1595-1599. doi: https://doi.org/10.12669/pjms.37.6-WIT.4848 This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

show abstract

“…Thus, highly efficient deep learning methods for automating the segmentation of fundus morphologies was discovered. A retinal image segmentation method is also proposed by Li et al, called the MAU-Net ( Li et al, 2020 ), that takes advantage of both modulated deformable convolutions and dual attention modules to realize vessel segmentation based on the U-net structure. Kromm & Rohr (2020) developed a novel deep learning method for vessel segmentation and centerline extraction of retinal blood vessels based on the Capsule network in combination with an Inception architecture.…”

Section: Introductionmentioning

confidence: 99%

“…A trimap is obtained via a bi-level thresholding of the score map using existing methods, which is instrumental in focusing the attention to the pixels of these unknown areas. Among these ANN methods, ( Yang et al, 2020 ; Kromm & Rohr, 2020 ; Adarsh et al, 2020 ; Guo et al, 2019 ; Yan, Yang & Cheng, 2019 ) have researched on multi-scale features, ( Li et al, 2020 ) has researched attention mechanisms, ( Ribeiro, Lopes & Silva, 2019 ) has researched ensemble strategy methods, Leopold et al (2017) has researched the dependency between pixels, and ( Zhao, Li & Cheng, 2020 ) has researched post-processing methods. These studies can improve the accuracy of segmentation models.…”

Section: Introductionmentioning

confidence: 99%

DBFU-Net: Double branch fusion U-Net with hard example weighting train strategy to segment retinal vessel

Huang

Zhong

Chen

et al. 2022

PeerJ Computer Science

View full text Add to dashboard Cite

Background Many fundus imaging modalities measure ocular changes. Automatic retinal vessel segmentation (RVS) is a significant fundus image-based method for the diagnosis of ophthalmologic diseases. However, precise vessel segmentation is a challenging task when detecting micro-changes in fundus images, e.g., tiny vessels, vessel edges, vessel lesions and optic disc edges. Methods In this paper, we will introduce a novel double branch fusion U-Net model that allows one of the branches to be trained by a weighting scheme that emphasizes harder examples to improve the overall segmentation performance. A new mask, we call a hard example mask, is needed for those examples that include a weighting strategy that is different from other methods. The method we propose extracts the hard example mask by morphology, meaning that the hard example mask does not need any rough segmentation model. To alleviate overfitting, we propose a random channel attention mechanism that is better than the drop-out method or the L2-regularization method in RVS. Results We have verified the proposed approach on the DRIVE, STARE and CHASE datasets to quantify the performance metrics. Compared to other existing approaches, using those dataset platforms, the proposed approach has competitive performance metrics. (DRIVE: F1-Score = 0.8289, G-Mean = 0.8995, AUC = 0.9811; STARE: F1-Score = 0.8501, G-Mean = 0.9198, AUC = 0.9892; CHASE: F1-Score = 0.8375, G-Mean = 0.9138, AUC = 0.9879). Discussion The segmentation results showed that DBFU-Net with RCA achieves competitive performance in three RVS datasets. Additionally, the proposed morphological-based extraction method for hard examples can reduce the computational cost. Finally, the random channel attention mechanism proposed in this paper has proven to be more effective than other regularization methods in the RVS task.

show abstract

MAU-Net: A Retinal Vessels Segmentation Method

Cited by 15 publications

References 8 publications

Extended U-net for Retinal Vessel Segmentation

Extended U-net for Retinal Vessel Segmentation

Construction and application of color fundus image segmentation algorithm based on Multi-Scale local combined global enhancement

DBFU-Net: Double branch fusion U-Net with hard example weighting train strategy to segment retinal vessel

Contact Info

Product

Resources

About