Image Quality Assessment of Diabetic Retinopathy Based on ADD-Net
Retinal image quality assessment (RIQA) is a precondition for diabetic retinopathy screening and diagnosis. Current RIQA network tend to place too much emphasis on performance improvement, ignoring the number of parameters and computational complexity of the network, and the existing publicly availa...
Main Authors: | , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
IEEE
2023-01-01
|
Series: | IEEE Access |
Subjects: | |
Online Access: | https://ieeexplore.ieee.org/document/10261978/ |
_version_ | 1797668061269983232 |
---|---|
author | Peiming Zhang Tong Hu Xiaoxiang Han Xiao Liu Jia Ni Jialin Shen Qiaohong Liu |
author_facet | Peiming Zhang Tong Hu Xiaoxiang Han Xiao Liu Jia Ni Jialin Shen Qiaohong Liu |
author_sort | Peiming Zhang |
collection | DOAJ |
description | Retinal image quality assessment (RIQA) is a precondition for diabetic retinopathy screening and diagnosis. Current RIQA network tend to place too much emphasis on performance improvement, ignoring the number of parameters and computational complexity of the network, and the existing publicly available datasets with unclear classification criteria and unbalanced number of different kinds of images, which affect the training of deep neural networks and large area screening of diabetic retinopathy. To solve these problems. A new quality criteria for retinal images is developed, and the Kaggle retinal image quality dataset is re-labeled and reclassified according to the criteria. And a high-performance lightweight RIQA network ADD-Net is proposed based on the DenseNet, which increases the perceptual field and improves pixel utilization by stacking dilated convolutions in DenseLayer, and incorporates attention mechanisms to further optimize the network performance. The proposed ADD-Net has a parametric number of 8.92M, a computational complexity of 2.76GMac, an accuracy of 97.03%, a precision of 97.06%, and a sensitivity of 98.31%, which are better than those of current mainstream methods. A quality criteria for retinal images that combines image quality features, retinal structural features, and DR lesion features was developed. The RIQA network ADD-Net with low number of parameters, less computational complexity and excellent performance is designed. |
first_indexed | 2024-03-11T20:23:37Z |
format | Article |
id | doaj.art-baa3890fef204974ad388ff34155968b |
institution | Directory Open Access Journal |
issn | 2169-3536 |
language | English |
last_indexed | 2024-03-11T20:23:37Z |
publishDate | 2023-01-01 |
publisher | IEEE |
record_format | Article |
series | IEEE Access |
spelling | doaj.art-baa3890fef204974ad388ff34155968b2023-10-02T23:00:18ZengIEEEIEEE Access2169-35362023-01-011110513010513910.1109/ACCESS.2023.331887610261978Image Quality Assessment of Diabetic Retinopathy Based on ADD-NetPeiming Zhang0Tong Hu1Xiaoxiang Han2Xiao Liu3Jia Ni4Jialin Shen5Qiaohong Liu6https://orcid.org/0009-0009-7637-8185School of Health Science and Engineering, University of Shanghai for Science and Technology, Shanghai, ChinaSchool of Health Science and Engineering, University of Shanghai for Science and Technology, Shanghai, ChinaSchool of Health Science and Engineering, University of Shanghai for Science and Technology, Shanghai, ChinaSchool of Health Science and Engineering, University of Shanghai for Science and Technology, Shanghai, ChinaSchool of Health Science and Engineering, University of Shanghai for Science and Technology, Shanghai, ChinaSchool of Health Science and Engineering, University of Shanghai for Science and Technology, Shanghai, ChinaMedical Instrumentation College, Shanghai University of Medicine and Health Sciences, Shanghai, ChinaRetinal image quality assessment (RIQA) is a precondition for diabetic retinopathy screening and diagnosis. Current RIQA network tend to place too much emphasis on performance improvement, ignoring the number of parameters and computational complexity of the network, and the existing publicly available datasets with unclear classification criteria and unbalanced number of different kinds of images, which affect the training of deep neural networks and large area screening of diabetic retinopathy. To solve these problems. A new quality criteria for retinal images is developed, and the Kaggle retinal image quality dataset is re-labeled and reclassified according to the criteria. And a high-performance lightweight RIQA network ADD-Net is proposed based on the DenseNet, which increases the perceptual field and improves pixel utilization by stacking dilated convolutions in DenseLayer, and incorporates attention mechanisms to further optimize the network performance. The proposed ADD-Net has a parametric number of 8.92M, a computational complexity of 2.76GMac, an accuracy of 97.03%, a precision of 97.06%, and a sensitivity of 98.31%, which are better than those of current mainstream methods. A quality criteria for retinal images that combines image quality features, retinal structural features, and DR lesion features was developed. The RIQA network ADD-Net with low number of parameters, less computational complexity and excellent performance is designed.https://ieeexplore.ieee.org/document/10261978/Attention mechanismdeep learningdilated convolutionDenseNetretinal image quality assessment (RIQA) |
spellingShingle | Peiming Zhang Tong Hu Xiaoxiang Han Xiao Liu Jia Ni Jialin Shen Qiaohong Liu Image Quality Assessment of Diabetic Retinopathy Based on ADD-Net IEEE Access Attention mechanism deep learning dilated convolution DenseNet retinal image quality assessment (RIQA) |
title | Image Quality Assessment of Diabetic Retinopathy Based on ADD-Net |
title_full | Image Quality Assessment of Diabetic Retinopathy Based on ADD-Net |
title_fullStr | Image Quality Assessment of Diabetic Retinopathy Based on ADD-Net |
title_full_unstemmed | Image Quality Assessment of Diabetic Retinopathy Based on ADD-Net |
title_short | Image Quality Assessment of Diabetic Retinopathy Based on ADD-Net |
title_sort | image quality assessment of diabetic retinopathy based on add net |
topic | Attention mechanism deep learning dilated convolution DenseNet retinal image quality assessment (RIQA) |
url | https://ieeexplore.ieee.org/document/10261978/ |
work_keys_str_mv | AT peimingzhang imagequalityassessmentofdiabeticretinopathybasedonaddnet AT tonghu imagequalityassessmentofdiabeticretinopathybasedonaddnet AT xiaoxianghan imagequalityassessmentofdiabeticretinopathybasedonaddnet AT xiaoliu imagequalityassessmentofdiabeticretinopathybasedonaddnet AT jiani imagequalityassessmentofdiabeticretinopathybasedonaddnet AT jialinshen imagequalityassessmentofdiabeticretinopathybasedonaddnet AT qiaohongliu imagequalityassessmentofdiabeticretinopathybasedonaddnet |