MSCNN-AM: A Multi-Scale Convolutional Neural Network With Attention Mechanisms for Retinal Vessel Segmentation
Automatic retinal vessel segmentation has drawn significant attention in early diagnosis and treatment of many diseases, such as diabetes, retinal diseases, and coronary heart disease. However, due to vessels exhibit variations in morphology and low contrast, it is still challenging to obtain accura...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
IEEE
2020-01-01
|
Series: | IEEE Access |
Subjects: | |
Online Access: | https://ieeexplore.ieee.org/document/9187192/ |
id |
doaj-4abcb67a33c14aa49fadd878d35e7351 |
---|---|
record_format |
Article |
spelling |
doaj-4abcb67a33c14aa49fadd878d35e73512021-03-30T03:55:35ZengIEEEIEEE Access2169-35362020-01-01816392616393610.1109/ACCESS.2020.30221779187192MSCNN-AM: A Multi-Scale Convolutional Neural Network With Attention Mechanisms for Retinal Vessel SegmentationQilong Fu0https://orcid.org/0000-0002-8829-3279Shuqiu Li1https://orcid.org/0000-0002-3331-3332Xin Wang2https://orcid.org/0000-0003-1537-2731College of Computer Science and Technology, Jilin University, Changchun, ChinaCollege of Computer Science and Technology, Jilin University, Changchun, ChinaCollege of Computer Science and Technology, Jilin University, Changchun, ChinaAutomatic retinal vessel segmentation has drawn significant attention in early diagnosis and treatment of many diseases, such as diabetes, retinal diseases, and coronary heart disease. However, due to vessels exhibit variations in morphology and low contrast, it is still challenging to obtain accurate segmentation results. In this paper, aiming at upgrading the accuracy and sensitivity of existing vessel segmentation methods, we propose a Multi-Scale Convolutional Neural Network with Attention Mechanisms (MSCNN-AM). For extraction of blood vessels at different scales, we introduce atrous separable convolutions with varying dilation rates, which could capture global and multi-scale vessel information better. Meanwhile, in order to reduce false-positive predictions for tiny vessel pixels, we also adopt attention mechanisms so that the proposed MSCNN-AM can pay more attention to retinal vessel pixels instead of background pixels. Because the green channel shows better vessel contrast and less noise than other channels in the RGB image, our proposed MSCNN-AM is trained and tested with green channel images only, excluding extra pre-processing and post-processing steps. The proposed method is evaluated on three public datasets, including DRIVE, STARE, and CHASE_DB1. In addition, we adopt six objective metrics to verify the performance of the MSCNN-AM, including sensitivity (Se), specificity (Sp), accuracy (Acc), F1-score, an area under a receiver operating characteristic curve (AUC-ROC), and an area under precision/recall curve (AUC-PR). Experimental results indicate that our proposed method outperforms most of the existing methods with a sensitivity of 0.8342/0.8412/0.8132 and an accuracy of 0.9555/0.9658/0.9644 on DRIVE, STARE, and CHASE_DB1 separately.https://ieeexplore.ieee.org/document/9187192/Retinal vessel segmentationconvolutional neural networkmulti-scale informationattention mechanism |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Qilong Fu Shuqiu Li Xin Wang |
spellingShingle |
Qilong Fu Shuqiu Li Xin Wang MSCNN-AM: A Multi-Scale Convolutional Neural Network With Attention Mechanisms for Retinal Vessel Segmentation IEEE Access Retinal vessel segmentation convolutional neural network multi-scale information attention mechanism |
author_facet |
Qilong Fu Shuqiu Li Xin Wang |
author_sort |
Qilong Fu |
title |
MSCNN-AM: A Multi-Scale Convolutional Neural Network With Attention Mechanisms for Retinal Vessel Segmentation |
title_short |
MSCNN-AM: A Multi-Scale Convolutional Neural Network With Attention Mechanisms for Retinal Vessel Segmentation |
title_full |
MSCNN-AM: A Multi-Scale Convolutional Neural Network With Attention Mechanisms for Retinal Vessel Segmentation |
title_fullStr |
MSCNN-AM: A Multi-Scale Convolutional Neural Network With Attention Mechanisms for Retinal Vessel Segmentation |
title_full_unstemmed |
MSCNN-AM: A Multi-Scale Convolutional Neural Network With Attention Mechanisms for Retinal Vessel Segmentation |
title_sort |
mscnn-am: a multi-scale convolutional neural network with attention mechanisms for retinal vessel segmentation |
publisher |
IEEE |
series |
IEEE Access |
issn |
2169-3536 |
publishDate |
2020-01-01 |
description |
Automatic retinal vessel segmentation has drawn significant attention in early diagnosis and treatment of many diseases, such as diabetes, retinal diseases, and coronary heart disease. However, due to vessels exhibit variations in morphology and low contrast, it is still challenging to obtain accurate segmentation results. In this paper, aiming at upgrading the accuracy and sensitivity of existing vessel segmentation methods, we propose a Multi-Scale Convolutional Neural Network with Attention Mechanisms (MSCNN-AM). For extraction of blood vessels at different scales, we introduce atrous separable convolutions with varying dilation rates, which could capture global and multi-scale vessel information better. Meanwhile, in order to reduce false-positive predictions for tiny vessel pixels, we also adopt attention mechanisms so that the proposed MSCNN-AM can pay more attention to retinal vessel pixels instead of background pixels. Because the green channel shows better vessel contrast and less noise than other channels in the RGB image, our proposed MSCNN-AM is trained and tested with green channel images only, excluding extra pre-processing and post-processing steps. The proposed method is evaluated on three public datasets, including DRIVE, STARE, and CHASE_DB1. In addition, we adopt six objective metrics to verify the performance of the MSCNN-AM, including sensitivity (Se), specificity (Sp), accuracy (Acc), F1-score, an area under a receiver operating characteristic curve (AUC-ROC), and an area under precision/recall curve (AUC-PR). Experimental results indicate that our proposed method outperforms most of the existing methods with a sensitivity of 0.8342/0.8412/0.8132 and an accuracy of 0.9555/0.9658/0.9644 on DRIVE, STARE, and CHASE_DB1 separately. |
topic |
Retinal vessel segmentation convolutional neural network multi-scale information attention mechanism |
url |
https://ieeexplore.ieee.org/document/9187192/ |
work_keys_str_mv |
AT qilongfu mscnnamamultiscaleconvolutionalneuralnetworkwithattentionmechanismsforretinalvesselsegmentation AT shuqiuli mscnnamamultiscaleconvolutionalneuralnetworkwithattentionmechanismsforretinalvesselsegmentation AT xinwang mscnnamamultiscaleconvolutionalneuralnetworkwithattentionmechanismsforretinalvesselsegmentation |
_version_ |
1724182615677730816 |