Wavelet-Attention Swin for Automatic Diabetic Retinopathy Classification

Diabetic retinopathy (DR) is a complication of diabetes that affects the eyes by damaging the blood vessels in the retina. High blood sugar levels can cause leakage or blockage of these vessels, leading to vision loss or blindness. Early detection of DR is crucial to prevent blindness, but manually...

Full description

Saved in:
Bibliographic Details
Main Authors: Rasha Ali Dihin, Ebtesam N. AlShemmary, Waleed A. M. Al-Jawher
Format: Article
Language:English
Published: University of Baghdad, College of Science for Women 2024-08-01
Series:مجلة بغداد للعلوم
Subjects:
Online Access:https://bsj.uobaghdad.edu.iq/index.php/BSJ/article/view/8565
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849411983479144448
author Rasha Ali Dihin
Ebtesam N. AlShemmary
Waleed A. M. Al-Jawher
author_facet Rasha Ali Dihin
Ebtesam N. AlShemmary
Waleed A. M. Al-Jawher
author_sort Rasha Ali Dihin
collection DOAJ
description Diabetic retinopathy (DR) is a complication of diabetes that affects the eyes by damaging the blood vessels in the retina. High blood sugar levels can cause leakage or blockage of these vessels, leading to vision loss or blindness. Early detection of DR is crucial to prevent blindness, but manually analyzing fundus images can be time-consuming, especially with a large number of images. Swin-Transformers have gained popularity in medical image analysis, reducing calculations and yielding improved results. This paper introduces the WT Attention-Db5 Block, which focuses attention on the high-frequency domain using Discrete Wavelet Transform (DWT). This block extracts detailed information from the high-frequency field while retaining essential low-frequency information. The study discusses findings from the 2019 Blindness Detection challenge (APTOS 2019 BD) held by the Asia Pacific Tele-Ophthalmology Society.The proposed WT-Swin model achieves significant improvements in classification accuracy. For Swin-T, the training and validation accuracies are 99.14% and 98.91%, respectively. For binary classification using Swin-B, the training accuracy is 99.01%, the validation accuracy is 99.18%, and the test accuracy is 98%. In multi-classification, the training and validation accuracies are 93.19% and 86.34%, respectively, while the test accuracy is 86%.In conclusion, early detection of DR is essential for preventing vision loss. The WT Attention-Db5 Block integrated into the WT-Swin model shows promising results in classification accuracy
format Article
id doaj-art-62a75bf8205141a8b49ec9bde01a0820
institution Kabale University
issn 2078-8665
2411-7986
language English
publishDate 2024-08-01
publisher University of Baghdad, College of Science for Women
record_format Article
series مجلة بغداد للعلوم
spelling doaj-art-62a75bf8205141a8b49ec9bde01a08202025-08-20T03:34:36ZengUniversity of Baghdad, College of Science for Womenمجلة بغداد للعلوم2078-86652411-79862024-08-0121810.21123/bsj.2024.8565Wavelet-Attention Swin for Automatic Diabetic Retinopathy ClassificationRasha Ali Dihin0Ebtesam N. AlShemmary 1https://orcid.org/0000-0001-7500-9702Waleed A. M. Al-Jawher2https://orcid.org/0000-0002-3660-7758Department of Computer Science, Faculty of Computer Science and Mathematics, University of Kufa, Kufa, Iraq.IT Research and Development Center, University of Kufa, Kufa, Iraq.Uruk University, Baghdad, Iraq. Diabetic retinopathy (DR) is a complication of diabetes that affects the eyes by damaging the blood vessels in the retina. High blood sugar levels can cause leakage or blockage of these vessels, leading to vision loss or blindness. Early detection of DR is crucial to prevent blindness, but manually analyzing fundus images can be time-consuming, especially with a large number of images. Swin-Transformers have gained popularity in medical image analysis, reducing calculations and yielding improved results. This paper introduces the WT Attention-Db5 Block, which focuses attention on the high-frequency domain using Discrete Wavelet Transform (DWT). This block extracts detailed information from the high-frequency field while retaining essential low-frequency information. The study discusses findings from the 2019 Blindness Detection challenge (APTOS 2019 BD) held by the Asia Pacific Tele-Ophthalmology Society.The proposed WT-Swin model achieves significant improvements in classification accuracy. For Swin-T, the training and validation accuracies are 99.14% and 98.91%, respectively. For binary classification using Swin-B, the training accuracy is 99.01%, the validation accuracy is 99.18%, and the test accuracy is 98%. In multi-classification, the training and validation accuracies are 93.19% and 86.34%, respectively, while the test accuracy is 86%.In conclusion, early detection of DR is essential for preventing vision loss. The WT Attention-Db5 Block integrated into the WT-Swin model shows promising results in classification accuracy https://bsj.uobaghdad.edu.iq/index.php/BSJ/article/view/8565APTOS Data Set, Diabetic Retinopathy, Swin-B, Swin-T, Wavelet-Attention
spellingShingle Rasha Ali Dihin
Ebtesam N. AlShemmary
Waleed A. M. Al-Jawher
Wavelet-Attention Swin for Automatic Diabetic Retinopathy Classification
مجلة بغداد للعلوم
APTOS Data Set, Diabetic Retinopathy, Swin-B, Swin-T, Wavelet-Attention
title Wavelet-Attention Swin for Automatic Diabetic Retinopathy Classification
title_full Wavelet-Attention Swin for Automatic Diabetic Retinopathy Classification
title_fullStr Wavelet-Attention Swin for Automatic Diabetic Retinopathy Classification
title_full_unstemmed Wavelet-Attention Swin for Automatic Diabetic Retinopathy Classification
title_short Wavelet-Attention Swin for Automatic Diabetic Retinopathy Classification
title_sort wavelet attention swin for automatic diabetic retinopathy classification
topic APTOS Data Set, Diabetic Retinopathy, Swin-B, Swin-T, Wavelet-Attention
url https://bsj.uobaghdad.edu.iq/index.php/BSJ/article/view/8565
work_keys_str_mv AT rashaalidihin waveletattentionswinforautomaticdiabeticretinopathyclassification
AT ebtesamnalshemmary waveletattentionswinforautomaticdiabeticretinopathyclassification
AT waleedamaljawher waveletattentionswinforautomaticdiabeticretinopathyclassification