• 文献检索
  • 文档翻译
  • 深度研究
  • 学术资讯
  • Suppr Zotero 插件Zotero 插件
  • 邀请有礼
  • 套餐&价格
  • 历史记录
应用&插件
Suppr Zotero 插件Zotero 插件浏览器插件Mac 客户端Windows 客户端微信小程序
定价
高级版会员购买积分包购买API积分包
服务
文献检索文档翻译深度研究API 文档MCP 服务
关于我们
关于 Suppr公司介绍联系我们用户协议隐私条款
关注我们

Suppr 超能文献

核心技术专利：CN118964589B侵权必究
粤ICP备2023148730 号-1Suppr @ 2026

文献检索

告别复杂PubMed语法，用中文像聊天一样搜索，搜遍4000万医学文献。AI智能推荐，让科研检索更轻松。

立即免费搜索

文件翻译

保留排版，准确专业，支持PDF/Word/PPT等文件格式，支持 12+语言互译。

免费翻译文档

深度研究

AI帮你快速写综述，25分钟生成高质量综述，智能提取关键信息，辅助科研写作。

立即免费体验

通过无监督深度学习实现基于血液的红外光谱的信息性表示

Toward Informative Representations of Blood-Based Infrared Spectra via Unsupervised Deep Learning.

作者信息

Wegner Corinna, Zarandy Zita I, Feiler Nico, Gigou Lea, Halenke Timo, Leopold-Kerschbaumer Niklas, Krusche Maik, Skibicka Weronika, Kepesidis Kosmas V

机构信息

Chair of Experimental Physics-Laser Physics, Ludwig-Maximilians-Universität München (LMU), Garching, Germany.

Laboratory for Attosecond Physics, Max Planck Institute of Quantum Optics (MPQ), Garching, Germany.

出版信息

J Biophotonics. 2025 Aug;18(8):e70011. doi: 10.1002/jbio.70011. Epub 2025 Mar 24.

DOI:10.1002/jbio.70011
PMID:40129053
原文链接:https://pmc.ncbi.nlm.nih.gov/articles/PMC12318647/
Abstract

This study explores using unsupervised deep learning to find a low-dimensional representation of infrared molecular fingerprints of human blood. We developed a fully convolutional denoising autoencoder to process Fourier transform infrared (FTIR) spectroscopy data, aiming to condense the spectra into a set of latent variables. By utilizing the autoencoder's bottleneck architecture and a custom loss function, we effectively reduced noise while retaining essential molecular information. This method improved lung cancer detection accuracy by 2.6 percentage points in a case-control study. The resulting latent space not only compacts spectral data, but also highlights variables linked to disease presence, offering potential for improving diagnostics. Trial Registration: German Clinical Trials Register (DRKS): DRKS00013217.

摘要

本研究探索使用无监督深度学习来寻找人类血液红外分子指纹的低维表示。我们开发了一种全卷积去噪自动编码器来处理傅里叶变换红外（FTIR）光谱数据，旨在将光谱压缩成一组潜在变量。通过利用自动编码器的瓶颈架构和自定义损失函数，我们在保留基本分子信息的同时有效降低了噪声。在一项病例对照研究中，该方法将肺癌检测准确率提高了2.6个百分点。所得的潜在空间不仅压缩了光谱数据，还突出了与疾病存在相关的变量，为改进诊断提供了潜力。试验注册：德国临床试验注册中心（DRKS）：DRKS00013217。

https://cdn.ncbi.nlm.nih.gov/pmc/blobs/1254/12318647/9c9f0d51b2a0/JBIO-18-e70011-g002.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/1254/12318647/ee35a2f241b0/JBIO-18-e70011-g005.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/1254/12318647/876257d29334/JBIO-18-e70011-g001.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/1254/12318647/3dba84b3cdfc/JBIO-18-e70011-g004.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/1254/12318647/88706c3b83b0/JBIO-18-e70011-g003.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/1254/12318647/9c9f0d51b2a0/JBIO-18-e70011-g002.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/1254/12318647/ee35a2f241b0/JBIO-18-e70011-g005.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/1254/12318647/876257d29334/JBIO-18-e70011-g001.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/1254/12318647/3dba84b3cdfc/JBIO-18-e70011-g004.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/1254/12318647/88706c3b83b0/JBIO-18-e70011-g003.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/1254/12318647/9c9f0d51b2a0/JBIO-18-e70011-g002.jpg

相似文献

1
Toward Informative Representations of Blood-Based Infrared Spectra via Unsupervised Deep Learning.通过无监督深度学习实现基于血液的红外光谱的信息性表示
J Biophotonics. 2025 Aug;18(8):e70011. doi: 10.1002/jbio.70011. Epub 2025 Mar 24.
2
Fourier Transform Infrared (FTIR) Spectroscopy as a Tool to Characterize Exercise and Physical Activity: A Systematic Review.傅里叶变换红外（FTIR）光谱作为表征运动和身体活动的工具：一项系统综述。
Sports Med. 2025 Feb;55(2):459-472. doi: 10.1007/s40279-024-02139-5. Epub 2024 Nov 20.
3
Enhancing microplastic classification through filter-interfered FTIR spectra using dimensionality reduction and deep learning in low-dimensional spaces.利用降维和深度学习在低维空间中通过滤光干涉傅里叶变换红外光谱增强微塑料分类
Mar Pollut Bull. 2025 Aug 16;221:118548. doi: 10.1016/j.marpolbul.2025.118548.
4
Safe-by-Design Strategies for Intranasal Drug Delivery Systems: Machine and Deep Learning Solutions to Differentiate Epithelial Tissues via Attenuated Total Reflection Fourier Transform Infrared Spectroscopy.鼻内给药系统的设计安全策略：通过衰减全反射傅里叶变换红外光谱法区分上皮组织的机器学习和深度学习解决方案
ACS Pharmacol Transl Sci. 2025 Feb 27;8(3):762-773. doi: 10.1021/acsptsci.4c00643. eCollection 2025 Mar 14.
5
A deep learning approach to direct immunofluorescence pattern recognition in autoimmune bullous diseases.深度学习方法在自身免疫性大疱性疾病中的直接免疫荧光模式识别。
Br J Dermatol. 2024 Jul 16;191(2):261-266. doi: 10.1093/bjd/ljae142.
6
Deep Learning for the Early Detection of Invasive Ductal Carcinoma in Histopathological Images: Convolutional Neural Network Approach With Transfer Learning.基于深度学习的组织病理学图像中浸润性导管癌早期检测：采用迁移学习的卷积神经网络方法
JMIR Form Res. 2025 Aug 21;9:e62996. doi: 10.2196/62996.
7
Label-free diagnosis of lung cancer by Fourier transform infrared microspectroscopy coupled with domain adversarial learning.基于傅里叶变换红外显微光谱结合域对抗学习的肺癌无标记诊断
Analyst. 2025 Aug 18;150(17):3800-3811. doi: 10.1039/d5an00216h.
8
Predicting reticuloruminal pH and subacute ruminal acidosis of individual cows using machine learning and Fourier-transform infrared spectroscopy milk analysis.使用机器学习和傅里叶变换红外光谱法进行牛奶分析来预测个体奶牛的瘤网胃pH值和亚急性瘤胃酸中毒。
J Dairy Sci. 2025 Aug;108(8):8606-8618. doi: 10.3168/jds.2024-25970. Epub 2025 Jun 9.
9
Evaluating robustly standardized explainable anomaly detection of implausible variables in cancer data.评估癌症数据中可疑变量的稳健标准化可解释异常检测。
J Am Med Inform Assoc. 2025 Apr 1;32(4):724-735. doi: 10.1093/jamia/ocaf011.
10
Use of early lactation milk Fourier transform mid-infrared spectroscopy and farm data to predict the calving-to-conception interval in Chinese dairy cows.利用早期泌乳牛奶傅里叶变换中红外光谱和农场数据预测中国奶牛的产犊至受孕间隔
J Dairy Sci. 2025 Jul 15. doi: 10.3168/jds.2025-26495.

本文引用的文献

1
Assessing lung cancer progression and survival with infrared spectroscopy of blood serum.通过血清红外光谱评估肺癌进展和生存率。
BMC Med. 2025 Feb 21;23(1):101. doi: 10.1186/s12916-025-03924-3.
2
Plasma infrared fingerprinting with machine learning enables single-measurement multi-phenotype health screening.结合机器学习的血浆红外指纹图谱技术可实现单次测量的多表型健康筛查。
Cell Rep Med. 2024 Jul 16;5(7):101625. doi: 10.1016/j.xcrm.2024.101625. Epub 2024 Jun 28.
3
Heat-vision based drone surveillance augmented by deep learning for critical industrial monitoring.
基于热视觉的无人机监测通过深度学习增强，用于关键工业监控。
Sci Rep. 2023 Dec 15;13(1):22291. doi: 10.1038/s41598-023-49589-x.
4
Scientific discovery in the age of artificial intelligence.人工智能时代的科学发现。
Nature. 2023 Aug;620(7972):47-60. doi: 10.1038/s41586-023-06221-2. Epub 2023 Aug 2.
5
Convolutional neural network-based respiration analysis of electrical activities of the diaphragm.基于卷积神经网络的膈神经电活动呼吸分析。
Sci Rep. 2022 Oct 5;12(1):16671. doi: 10.1038/s41598-022-21165-9.
6
Five ways deep learning has transformed image analysis.深度学习改变图像分析的五种方式。
Nature. 2022 Sep;609(7928):864-866. doi: 10.1038/d41586-022-02964-6.
7
Learning disentangled representations in the imaging domain.在成像领域中学习解缠表示。
Med Image Anal. 2022 Aug;80:102516. doi: 10.1016/j.media.2022.102516. Epub 2022 Jun 17.
8
Breast-cancer detection using blood-based infrared molecular fingerprints.基于血液的红外分子指纹图谱检测乳腺癌。
BMC Cancer. 2021 Dec 2;21(1):1287. doi: 10.1186/s12885-021-09017-7.
9
Infrared molecular fingerprinting of blood-based liquid biopsies for the detection of cancer.基于血液的液体活检的红外分子指纹图谱用于癌症检测。
Elife. 2021 Oct 26;10:e68758. doi: 10.7554/eLife.68758.
10
Biomarkers and Lung Cancer Early Detection: State of the Art.生物标志物与肺癌早期检测：现状
Cancers (Basel). 2021 Aug 3;13(15):3919. doi: 10.3390/cancers13153919.