Enhancing diabetic retinopathy and macular edema detection through multi scale feature fusion using deep learning model.

IF 2.4 3区医学 Q2 OPHTHALMOLOGY Graefe’s Archive for Clinical and Experimental Ophthalmology Pub Date : 2024-12-16 DOI:10.1007/s00417-024-06687-4

Gowri L, Haris R, Sumathi M, S P Raja

{"title":"Enhancing diabetic retinopathy and macular edema detection through multi scale feature fusion using deep learning model.","authors":"Gowri L, Haris R, Sumathi M, S P Raja","doi":"10.1007/s00417-024-06687-4","DOIUrl":null,"url":null,"abstract":"Background: This work tackles the growing problem of early identification of diabetic retinopathy and diabetic macular edema. The deep neural network design utilizes multi-scale feature fusion to improve automated diagnostic accuracy. Methods This approach uses convolutional neural networks (CNN) and is designed to combine higher-level semantic inputs with low-level textural characteristics. The contextual and localized abstract representations that complement each other are combined via a unique fusion technique.Results: Use the MESSIDOR dataset, which comprises retinal images labeled with pathological annotations, for model training and validation to ensure robust algorithm development. The suggested model shows a 98% general precision and good performance in diabetic retinopathy. This model achieves an impressive nearly 100% exactness for diabetic macular edema, with particularly high accuracy (0.99).Conclusion: Consistent performance increases the likelihood that the vision will be upheld through public screening and extensive clinical integration.","PeriodicalId":12795,"journal":{"name":"Graefe’s Archive for Clinical and Experimental Ophthalmology","volume":" ","pages":""},"PeriodicalIF":2.4000,"publicationDate":"2024-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Graefe’s Archive for Clinical and Experimental Ophthalmology","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1007/s00417-024-06687-4","RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"OPHTHALMOLOGY","Score":null,"Total":0}

引用次数: 0

Abstract

Background: This work tackles the growing problem of early identification of diabetic retinopathy and diabetic macular edema. The deep neural network design utilizes multi-scale feature fusion to improve automated diagnostic accuracy. Methods This approach uses convolutional neural networks (CNN) and is designed to combine higher-level semantic inputs with low-level textural characteristics. The contextual and localized abstract representations that complement each other are combined via a unique fusion technique.

Results: Use the MESSIDOR dataset, which comprises retinal images labeled with pathological annotations, for model training and validation to ensure robust algorithm development. The suggested model shows a 98% general precision and good performance in diabetic retinopathy. This model achieves an impressive nearly 100% exactness for diabetic macular edema, with particularly high accuracy (0.99).

Conclusion: Consistent performance increases the likelihood that the vision will be upheld through public screening and extensive clinical integration.

查看原文

微信好友朋友圈 QQ好友复制链接

本刊更多论文

利用深度学习模型，通过多尺度特征融合增强糖尿病视网膜病变和黄斑水肿检测。

背景：这项研究解决了糖尿病视网膜病变和糖尿病黄斑水肿早期识别这一日益严重的问题。深度神经网络设计利用多尺度特征融合来提高自动诊断的准确性。方法这种方法使用卷积神经网络（CNN），旨在将高层语义输入与低层纹理特征相结合。通过独特的融合技术将互为补充的上下文和局部抽象表征结合起来：使用 MESSIDOR 数据集（由标有病理注释的视网膜图像组成）进行模型训练和验证，以确保算法开发的稳健性。建议的模型在糖尿病视网膜病变中显示出 98% 的一般精确度和良好的性能。该模型在糖尿病黄斑水肿方面的精确度接近 100%，尤其是精确度高达 0.99，令人印象深刻：结论：通过公共筛查和广泛的临床整合，一致的性能提高了维护视力的可能性。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文去求助

来源期刊

Graefe’s Archive for Clinical and Experimental Ophthalmology 医学-眼科学

CiteScore

5.40

自引率

7.40%

发文量

398

审稿时长

3 months

期刊介绍： Graefe''s Archive for Clinical and Experimental Ophthalmology is a distinguished international journal that presents original clinical reports and clini-cally relevant experimental studies. Founded in 1854 by Albrecht von Graefe to serve as a source of useful clinical information and a stimulus for discussion, the journal has published articles by leading ophthalmologists and vision research scientists for more than a century. With peer review by an international Editorial Board and prompt English-language publication, Graefe''s Archive provides rapid dissemination of clinical and clinically related experimental information.