This Microtubule Does Not Exist: Super-Resolution Microscopy Image Generation by a Diffusion Model.

IF 10.7 2区 材料科学 Q1 CHEMISTRY, PHYSICAL
Alon Saguy, Tav Nahimov, Maia Lehrman, Estibaliz Gómez-de-Mariscal, Iván Hidalgo-Cenalmor, Onit Alalouf, Ashwin Balakrishnan, Mike Heilemann, Ricardo Henriques, Yoav Shechtman
{"title":"This Microtubule Does Not Exist: Super-Resolution Microscopy Image Generation by a Diffusion Model.","authors":"Alon Saguy, Tav Nahimov, Maia Lehrman, Estibaliz Gómez-de-Mariscal, Iván Hidalgo-Cenalmor, Onit Alalouf, Ashwin Balakrishnan, Mike Heilemann, Ricardo Henriques, Yoav Shechtman","doi":"10.1002/smtd.202400672","DOIUrl":null,"url":null,"abstract":"<p><p>Generative models, such as diffusion models, have made significant advancements in recent years, enabling the synthesis of high-quality realistic data across various domains. Here, the adaptation and training of a diffusion model on super-resolution microscopy images are explored. It is shown that the generated images resemble experimental images, and that the generation process does not exhibit a large degree of memorization from existing images in the training set. To demonstrate the usefulness of the generative model for data augmentation, the performance of a deep learning-based single-image super-resolution (SISR) method trained using generated high-resolution data is compared against training using experimental images alone, or images generated by mathematical modeling. Using a few experimental images, the reconstruction quality and the spatial resolution of the reconstructed images are improved, showcasing the potential of diffusion model image generation for overcoming the limitations accompanying the collection and annotation of microscopy images. Finally, the pipeline is made publicly available, runnable online, and user-friendly to enable researchers to generate their own synthetic microscopy data. This work demonstrates the potential contribution of generative diffusion models for microscopy tasks and paves the way for their future application in this field.</p>","PeriodicalId":229,"journal":{"name":"Small Methods","volume":" ","pages":"e2400672"},"PeriodicalIF":10.7000,"publicationDate":"2024-10-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Small Methods","FirstCategoryId":"88","ListUrlMain":"https://doi.org/10.1002/smtd.202400672","RegionNum":2,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"CHEMISTRY, PHYSICAL","Score":null,"Total":0}
引用次数: 0

Abstract

Generative models, such as diffusion models, have made significant advancements in recent years, enabling the synthesis of high-quality realistic data across various domains. Here, the adaptation and training of a diffusion model on super-resolution microscopy images are explored. It is shown that the generated images resemble experimental images, and that the generation process does not exhibit a large degree of memorization from existing images in the training set. To demonstrate the usefulness of the generative model for data augmentation, the performance of a deep learning-based single-image super-resolution (SISR) method trained using generated high-resolution data is compared against training using experimental images alone, or images generated by mathematical modeling. Using a few experimental images, the reconstruction quality and the spatial resolution of the reconstructed images are improved, showcasing the potential of diffusion model image generation for overcoming the limitations accompanying the collection and annotation of microscopy images. Finally, the pipeline is made publicly available, runnable online, and user-friendly to enable researchers to generate their own synthetic microscopy data. This work demonstrates the potential contribution of generative diffusion models for microscopy tasks and paves the way for their future application in this field.

这条微管并不存在:通过扩散模型生成超级分辨率显微镜图像
近年来,生成模型(如扩散模型)取得了长足的进步,能够合成各种领域的高质量真实数据。本文探讨了在超分辨率显微镜图像上对扩散模型进行适配和训练的问题。结果表明,生成的图像与实验图像非常相似,而且生成过程并没有表现出对训练集中现有图像的高度记忆。为了证明生成模型在数据增强方面的实用性,我们将使用生成的高分辨率数据训练的基于深度学习的单图像超分辨率(SISR)方法的性能与仅使用实验图像或通过数学建模生成的图像进行的训练进行了比较。通过使用少量实验图像,重建图像的重建质量和空间分辨率都得到了提高,从而展示了扩散模型图像生成在克服显微图像收集和注释限制方面的潜力。最后,我们公开了这一可在线运行、用户友好的管道,使研究人员能够生成自己的合成显微镜数据。这项工作证明了生成式扩散模型对显微镜任务的潜在贡献,并为其在该领域的未来应用铺平了道路。
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
Small Methods
Small Methods Materials Science-General Materials Science
CiteScore
17.40
自引率
1.60%
发文量
347
期刊介绍: Small Methods is a multidisciplinary journal that publishes groundbreaking research on methods relevant to nano- and microscale research. It welcomes contributions from the fields of materials science, biomedical science, chemistry, and physics, showcasing the latest advancements in experimental techniques. With a notable 2022 Impact Factor of 12.4 (Journal Citation Reports, Clarivate Analytics, 2023), Small Methods is recognized for its significant impact on the scientific community. The online ISSN for Small Methods is 2366-9608.
×
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