Improved diabetic retinopathy severity classification using squeeze-and-excitation and sparse light weight multi-level attention u-net with transfer learning from xception.
Sachin Bhandari, Sunil Pathak, Sonal Amit Jain, Basant Agarwal
{"title":"Improved diabetic retinopathy severity classification using squeeze-and-excitation and sparse light weight multi-level attention u-net with transfer learning from xception.","authors":"Sachin Bhandari, Sunil Pathak, Sonal Amit Jain, Basant Agarwal","doi":"10.1007/s00592-024-02341-x","DOIUrl":null,"url":null,"abstract":"<p><strong>Aims: </strong>Diabetic Retinopathy (DR) is a significant cause of vision loss in diabetic patients, making early detection and accurate severity classification essential for effective management and prevention. This study aims to develop an enhanced DR severity classification approach using advanced model architectures and transfer learning to improve diagnostic accuracy and support better patient care.</p><p><strong>Methods: </strong>We propose a novel model, Xception Squeeze-and-Excitation Sparse Lightweight Multi-Level Attention U-Net (XceSE_SparseLwMLA-UNet), designed to classify DR severity using fundus images from the Messidor 1 and Messidor 2 datasets. The XceSE_SparseLwMLA-UNet integrates several advanced mechanisms: the Squeeze-and-Excitation (SE) mechanism for adaptive feature recalibration, the Sparse Lightweight Multi-Level Attention (SparseLwMLA) mechanism for effective contextual information integration, and transfer learning from the Xception architecture to enhance feature extraction capabilities. The SE mechanism refines channel-wise feature responses, while SparseLwMLA enhances the model's ability to identify complex DR patterns. Transfer learning utilizes pre-trained weights from Xception to improve generalization across DR severity levels.</p><p><strong>Results: </strong>The proposed XceSE_SparseLwMLA-UNet model demonstrates superior performance in DR severity classification, achieving higher accuracy and improved multi-class F1 scores compared to existing models. The model's color-coded segmentation outputs offer interpretable visual representations, aiding medical professionals in assessing DR severity levels.</p><p><strong>Conclusions: </strong>The XceSE_SparseLwMLA-UNet model shows promise for advancing early DR diagnosis and management by enhancing classification accuracy and providing valuable visual insights. Its integration of advanced architectural features and transfer learning contributes to better patient care and improved visual health outcomes.</p>","PeriodicalId":6921,"journal":{"name":"Acta Diabetologica","volume":null,"pages":null},"PeriodicalIF":3.1000,"publicationDate":"2024-07-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Acta Diabetologica","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1007/s00592-024-02341-x","RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"ENDOCRINOLOGY & METABOLISM","Score":null,"Total":0}
引用次数: 0
Abstract
Aims: Diabetic Retinopathy (DR) is a significant cause of vision loss in diabetic patients, making early detection and accurate severity classification essential for effective management and prevention. This study aims to develop an enhanced DR severity classification approach using advanced model architectures and transfer learning to improve diagnostic accuracy and support better patient care.
Methods: We propose a novel model, Xception Squeeze-and-Excitation Sparse Lightweight Multi-Level Attention U-Net (XceSE_SparseLwMLA-UNet), designed to classify DR severity using fundus images from the Messidor 1 and Messidor 2 datasets. The XceSE_SparseLwMLA-UNet integrates several advanced mechanisms: the Squeeze-and-Excitation (SE) mechanism for adaptive feature recalibration, the Sparse Lightweight Multi-Level Attention (SparseLwMLA) mechanism for effective contextual information integration, and transfer learning from the Xception architecture to enhance feature extraction capabilities. The SE mechanism refines channel-wise feature responses, while SparseLwMLA enhances the model's ability to identify complex DR patterns. Transfer learning utilizes pre-trained weights from Xception to improve generalization across DR severity levels.
Results: The proposed XceSE_SparseLwMLA-UNet model demonstrates superior performance in DR severity classification, achieving higher accuracy and improved multi-class F1 scores compared to existing models. The model's color-coded segmentation outputs offer interpretable visual representations, aiding medical professionals in assessing DR severity levels.
Conclusions: The XceSE_SparseLwMLA-UNet model shows promise for advancing early DR diagnosis and management by enhancing classification accuracy and providing valuable visual insights. Its integration of advanced architectural features and transfer learning contributes to better patient care and improved visual health outcomes.
期刊介绍:
Acta Diabetologica is a journal that publishes reports of experimental and clinical research on diabetes mellitus and related metabolic diseases. Original contributions on biochemical, physiological, pathophysiological and clinical aspects of research on diabetes and metabolic diseases are welcome. Reports are published in the form of original articles, short communications and letters to the editor. Invited reviews and editorials are also published. A Methodology forum, which publishes contributions on methodological aspects of diabetes in vivo and in vitro, is also available. The Editor-in-chief will be pleased to consider articles describing new techniques (e.g., new transplantation methods, metabolic models), of innovative importance in the field of diabetes/metabolism. Finally, workshop reports are also welcome in Acta Diabetologica.