Gated Skip-Connection Network with Adaptive Upsampling for Retinal Vessel Segmentation
Segmentation of retinal vessels is a critical step for the diagnosis of some fundus diseases. <i><b>Methods:</b></i> To further enhance the performance of vessel segmentation, we propose a method based on a gated skip-connection network with adaptive upsampling (GSAU-Net). In...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2021-09-01
|
Series: | Sensors |
Subjects: | |
Online Access: | https://www.mdpi.com/1424-8220/21/18/6177 |
_version_ | 1797517180002107392 |
---|---|
author | Yun Jiang Huixia Yao Shengxin Tao Jing Liang |
author_facet | Yun Jiang Huixia Yao Shengxin Tao Jing Liang |
author_sort | Yun Jiang |
collection | DOAJ |
description | Segmentation of retinal vessels is a critical step for the diagnosis of some fundus diseases. <i><b>Methods:</b></i> To further enhance the performance of vessel segmentation, we propose a method based on a gated skip-connection network with adaptive upsampling (GSAU-Net). In GSAU-Net, a novel skip-connection with gating is first utilized in the extension path, which facilitates the flow of information from the encoder to the decoder. Specifically, we used the gated skip-connection between the encoder and decoder to gate the lower-level information from the encoder. In the decoding phase, we used an adaptive upsampling to replace the bilinear interpolation, which recovers feature maps from the decoder to obtain the pixelwise prediction. Finally, we validated our method on the DRIVE, CHASE, and STARE datasets. <i><b>Results:</b></i> The experimental results showed that our proposed method outperformed some existing methods, such as DeepVessel, AG-Net, and IterNet, in terms of accuracy, F-measure, and AUC<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><msub><mrow></mrow><mi>R</mi></msub><msub><mrow></mrow><mi>O</mi></msub><msub><mrow></mrow><mi>C</mi></msub></mrow></semantics></math></inline-formula>. The proposed method achieved a vessel segmentation F-measure of 83.13%, 81.40%, and 84.84% on the DRIVE, CHASE, and STARE datasets, respectively. |
first_indexed | 2024-03-10T07:13:09Z |
format | Article |
id | doaj.art-0bec1b3194ed4955b342922da7357795 |
institution | Directory Open Access Journal |
issn | 1424-8220 |
language | English |
last_indexed | 2024-03-10T07:13:09Z |
publishDate | 2021-09-01 |
publisher | MDPI AG |
record_format | Article |
series | Sensors |
spelling | doaj.art-0bec1b3194ed4955b342922da73577952023-11-22T15:12:51ZengMDPI AGSensors1424-82202021-09-012118617710.3390/s21186177Gated Skip-Connection Network with Adaptive Upsampling for Retinal Vessel SegmentationYun Jiang0Huixia Yao1Shengxin Tao2Jing Liang3College of Computer Science and Engineering, Northwest Normal University, Lanzhou 730070, ChinaCollege of Computer Science and Engineering, Northwest Normal University, Lanzhou 730070, ChinaCollege of Computer Science and Engineering, Northwest Normal University, Lanzhou 730070, ChinaCollege of Computer Science and Engineering, Northwest Normal University, Lanzhou 730070, ChinaSegmentation of retinal vessels is a critical step for the diagnosis of some fundus diseases. <i><b>Methods:</b></i> To further enhance the performance of vessel segmentation, we propose a method based on a gated skip-connection network with adaptive upsampling (GSAU-Net). In GSAU-Net, a novel skip-connection with gating is first utilized in the extension path, which facilitates the flow of information from the encoder to the decoder. Specifically, we used the gated skip-connection between the encoder and decoder to gate the lower-level information from the encoder. In the decoding phase, we used an adaptive upsampling to replace the bilinear interpolation, which recovers feature maps from the decoder to obtain the pixelwise prediction. Finally, we validated our method on the DRIVE, CHASE, and STARE datasets. <i><b>Results:</b></i> The experimental results showed that our proposed method outperformed some existing methods, such as DeepVessel, AG-Net, and IterNet, in terms of accuracy, F-measure, and AUC<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><msub><mrow></mrow><mi>R</mi></msub><msub><mrow></mrow><mi>O</mi></msub><msub><mrow></mrow><mi>C</mi></msub></mrow></semantics></math></inline-formula>. The proposed method achieved a vessel segmentation F-measure of 83.13%, 81.40%, and 84.84% on the DRIVE, CHASE, and STARE datasets, respectively.https://www.mdpi.com/1424-8220/21/18/6177deep convolutional neural workretinal vessel segmentationgating mechanismskip-connectionadaptive upsampling |
spellingShingle | Yun Jiang Huixia Yao Shengxin Tao Jing Liang Gated Skip-Connection Network with Adaptive Upsampling for Retinal Vessel Segmentation Sensors deep convolutional neural work retinal vessel segmentation gating mechanism skip-connection adaptive upsampling |
title | Gated Skip-Connection Network with Adaptive Upsampling for Retinal Vessel Segmentation |
title_full | Gated Skip-Connection Network with Adaptive Upsampling for Retinal Vessel Segmentation |
title_fullStr | Gated Skip-Connection Network with Adaptive Upsampling for Retinal Vessel Segmentation |
title_full_unstemmed | Gated Skip-Connection Network with Adaptive Upsampling for Retinal Vessel Segmentation |
title_short | Gated Skip-Connection Network with Adaptive Upsampling for Retinal Vessel Segmentation |
title_sort | gated skip connection network with adaptive upsampling for retinal vessel segmentation |
topic | deep convolutional neural work retinal vessel segmentation gating mechanism skip-connection adaptive upsampling |
url | https://www.mdpi.com/1424-8220/21/18/6177 |
work_keys_str_mv | AT yunjiang gatedskipconnectionnetworkwithadaptiveupsamplingforretinalvesselsegmentation AT huixiayao gatedskipconnectionnetworkwithadaptiveupsamplingforretinalvesselsegmentation AT shengxintao gatedskipconnectionnetworkwithadaptiveupsamplingforretinalvesselsegmentation AT jingliang gatedskipconnectionnetworkwithadaptiveupsamplingforretinalvesselsegmentation |