Jove
Visualize
联系我们
JoVE
x logofacebook logolinkedin logoyoutube logo
关于 JoVE
概览领导团队博客JoVE 帮助中心
作者
出版流程编辑委员会范围与政策同行评审常见问题投稿
图书馆员
用户评价订阅访问资源图书馆顾问委员会常见问题
研究
JoVE JournalMethods CollectionsJoVE Encyclopedia of Experiments存档
教育
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab Manual教师资源中心教师网站
使用条款与条件
隐私政策
政策

相关概念视频

您也可能阅读

相关文章

通过共同作者、期刊和引用图与本文相关的文章。

排序
Same author

Modifier-Sensitive Phenotypic Divergence in XMEN Disease (MAGT1 Deficiency): Neurodegenerative and Immuno-Hematologic Trajectories.

Journal of clinical medicine·2026
Same author

Should We Adopt Increased Dilutions for Indirect Immunofluorescence in Pediatric Anti-Centromere Antibody Testing? Insights from a Three-Year Retrospective Study.

Children (Basel, Switzerland)·2025
Same author

COVID-19 Antibody Levels among Various Vaccination Groups, One-Year Antibody Follow-Up in Two University Hospitals from Western and Central Turkey.

Vaccines·2024
Same author

Efficacy of coronavirus disease 2019 vaccines in patients with rheumatic diseases.

Archives of rheumatology·2023

相关实验视频

Updated: Jan 14, 2026

Simultaneous Distinction of Monospecific and Mixed DFS70 Patterns During ANA Screening with a Novel HEp-2 ELITE/DFS70 Knockout Substrate
10:05

Simultaneous Distinction of Monospecific and Mixed DFS70 Patterns During ANA Screening with a Novel HEp-2 ELITE/DFS70 Knockout Substrate

Published on: January 17, 2018

37.1K

通过EfficientNet架构和深度学习来识别IgA类内体抗体对子肝基质的同等结合模式.

Mehmet Soylu1, Ahmet Selman Bozkir2

  • 1Department of Medical Microbiology, Ege University, İzmir, Turkey.

PeerJ
|October 20, 2025
PubMed
概括

深度学习模型,特别是EfficientNetV2-S,在解释免疫球蛋白A (IgA) 固体抗体 (EMA) 测试以诊断乳病时表现出高准确性. 这些人工智能工具为实现更客观,更有效的诊断过程提供了途径.

关键词:
结核病的诊断是什么计算机视觉 计算机视觉 计算机视觉检测IgA内体抗体的测试.机器学习 机器学习

更多相关视频

Antibody Binding Specificity for Kappa (Vκ) Light Chain-containing Human (IgM) Antibodies: Polysialic Acid (PSA) Attached to NCAM as a Case Study
11:10

Antibody Binding Specificity for Kappa (Vκ) Light Chain-containing Human (IgM) Antibodies: Polysialic Acid (PSA) Attached to NCAM as a Case Study

Published on: June 29, 2016

14.5K
Identification of Mouse and Human Antibody Repertoires by Next-Generation Sequencing
08:51

Identification of Mouse and Human Antibody Repertoires by Next-Generation Sequencing

Published on: March 15, 2019

13.0K

相关实验视频

Last Updated: Jan 14, 2026

Simultaneous Distinction of Monospecific and Mixed DFS70 Patterns During ANA Screening with a Novel HEp-2 ELITE/DFS70 Knockout Substrate
10:05

Simultaneous Distinction of Monospecific and Mixed DFS70 Patterns During ANA Screening with a Novel HEp-2 ELITE/DFS70 Knockout Substrate

Published on: January 17, 2018

37.1K
Antibody Binding Specificity for Kappa (Vκ) Light Chain-containing Human (IgM) Antibodies: Polysialic Acid (PSA) Attached to NCAM as a Case Study
11:10

Antibody Binding Specificity for Kappa (Vκ) Light Chain-containing Human (IgM) Antibodies: Polysialic Acid (PSA) Attached to NCAM as a Case Study

Published on: June 29, 2016

14.5K
Identification of Mouse and Human Antibody Repertoires by Next-Generation Sequencing
08:51

Identification of Mouse and Human Antibody Repertoires by Next-Generation Sequencing

Published on: March 15, 2019

13.0K

科学领域:

  • 医学诊断 医学诊断 医学诊断
  • 人工智能在医学中的应用
  • 免疫学 免疫学 免疫学

背景情况:

  • 免疫球蛋白A (IgA) 固体抗体 (EMA) 测试对于诊断腹腔疾病至关重要.
  • 目前EMA的测试解释是主观和劳动密集型的,需要专家的人类分析.
  • 使用深度学习自动化EMA测试解释可以提高诊断效率和客观性.

研究的目的:

  • 评估EfficientNet和EfficientNetV2深度学习架构的性能,用于自动化IGA EMA测试解释.
  • 评估对二元,三级和四级场景的分类准确性,包括灰色区域的情况.
  • 探索可解释AI (HiRes-CAM) 在理解EMA-eq测试的深度学习模型决策方面的潜力.

主要方法:

  • 使用EfficientNet和EfficientNetV2架构进行IGA EMA等效 (EMA-eq) 测试的图像分类.
  • 在368个临床样本上训练并测试模型,通过不同的分类任务 (二进制,三类,四类).
  • 使用HiRes-CAM可视化和解释深度学习模型的解释过程.

主要成果:

  • EfficientNetV2-S实现了高精度:99.37% (二进制),95.28% (三进制) 和86.98% (四进制).
  • 与预期相反,中型深度学习架构的表现优于较大的架构.
  • 更高的输入分辨率 (640x640) 和EfficientNet-V2中的架构创新有助于提高性能.

结论:

  • 深度学习模型可以在解释IGA EMA-eq测试时达到专家级别的性能.
  • 自动翻译提供了一种更标准化,更有效,更客观的方法来诊断腹腔疾病.
  • 这项技术有可能减少专业医疗人员的工作量.