Unified deep learning model for predicting fundus fluorescein angiography image from fundus structure image

IF 2.3 3区 医学 Q2 OPTICS
Yiwei Chen, Yi He, Hong Ye, Lina Xing, Xin Zhang, Guohua Shi
{"title":"Unified deep learning model for predicting fundus fluorescein angiography image from fundus structure image","authors":"Yiwei Chen, Yi He, Hong Ye, Lina Xing, Xin Zhang, Guohua Shi","doi":"10.1142/s1793545824500032","DOIUrl":null,"url":null,"abstract":"<p>The prediction of fundus fluorescein angiography (FFA) images from fundus structural images is a cutting-edge research topic in ophthalmological image processing. Prediction comprises estimating FFA from fundus camera imaging, single-phase FFA from scanning laser ophthalmoscopy (SLO), and three-phase FFA also from SLO. Although many deep learning models are available, a single model can only perform one or two of these prediction tasks. To accomplish three prediction tasks using a unified method, we propose a unified deep learning model for predicting FFA images from fundus structure images using a supervised generative adversarial network. The three prediction tasks are processed as follows: data preparation, network training under FFA supervision, and FFA image prediction from fundus structure images on a test set. By comparing the FFA images predicted by our model, pix2pix, and CycleGAN, we demonstrate the remarkable progress achieved by our proposal. The high performance of our model is validated in terms of the peak signal-to-noise ratio, structural similarity index, and mean squared error.</p>","PeriodicalId":16248,"journal":{"name":"Journal of Innovative Optical Health Sciences","volume":null,"pages":null},"PeriodicalIF":2.3000,"publicationDate":"2024-03-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Innovative Optical Health Sciences","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1142/s1793545824500032","RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"OPTICS","Score":null,"Total":0}
引用次数: 0

Abstract

The prediction of fundus fluorescein angiography (FFA) images from fundus structural images is a cutting-edge research topic in ophthalmological image processing. Prediction comprises estimating FFA from fundus camera imaging, single-phase FFA from scanning laser ophthalmoscopy (SLO), and three-phase FFA also from SLO. Although many deep learning models are available, a single model can only perform one or two of these prediction tasks. To accomplish three prediction tasks using a unified method, we propose a unified deep learning model for predicting FFA images from fundus structure images using a supervised generative adversarial network. The three prediction tasks are processed as follows: data preparation, network training under FFA supervision, and FFA image prediction from fundus structure images on a test set. By comparing the FFA images predicted by our model, pix2pix, and CycleGAN, we demonstrate the remarkable progress achieved by our proposal. The high performance of our model is validated in terms of the peak signal-to-noise ratio, structural similarity index, and mean squared error.

从眼底结构图像预测眼底荧光素血管造影图像的统一深度学习模型
从眼底结构图像预测眼底荧光素血管造影(FFA)图像是眼科图像处理领域的前沿研究课题。预测包括通过眼底照相机成像估算 FFA、通过扫描激光眼底镜(SLO)估算单相 FFA 和通过 SLO 估算三相 FFA。虽然目前有许多深度学习模型,但单一模型只能完成其中的一两项预测任务。为了用一种统一的方法完成三项预测任务,我们提出了一种统一的深度学习模型,利用有监督的生成对抗网络从眼底结构图像中预测 FFA 图像。三项预测任务的处理过程如下:数据准备、FFA 监督下的网络训练以及在测试集上从眼底结构图像预测 FFA 图像。通过比较我们的模型、pix2pix 和 CycleGAN 预测的 FFA 图像,我们证明了我们的建议所取得的显著进步。我们模型的高性能在峰值信噪比、结构相似性指数和均方误差方面都得到了验证。
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 求助全文
来源期刊
Journal of Innovative Optical Health Sciences
Journal of Innovative Optical Health Sciences OPTICS-RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING
CiteScore
4.50
自引率
20.00%
发文量
69
审稿时长
>12 weeks
期刊介绍: JIOHS serves as an international forum for the publication of the latest developments in all areas of photonics in biology and medicine. JIOHS will consider for publication original papers in all disciplines of photonics in biology and medicine, including but not limited to: -Photonic therapeutics and diagnostics- Optical clinical technologies and systems- Tissue optics- Laser-tissue interaction and tissue engineering- Biomedical spectroscopy- Advanced microscopy and imaging- Nanobiophotonics and optical molecular imaging- Multimodal and hybrid biomedical imaging- Micro/nanofabrication- Medical microsystems- Optical coherence tomography- Photodynamic therapy. JIOHS provides a vehicle to help professionals, graduates, engineers, academics and researchers working in the field of intelligent photonics in biology and medicine to disseminate information on the state-of-the-art technique.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信