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

The Retina01:32

The Retina

74.3K
The retina is a layer of nervous tissue at the back of the eye that transduces light into neural signals. This process, called phototransduction, is carried out by rod and cone photoreceptor cells in the back of the retina.
74.3K

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

An ensemble-based 3D residual network for the classification of Alzheimer's disease.

PloS one·2025
Same author

Assessing a respiratory toxic infectious bronchitis virus (IBV) strain: isolation, identification, pathogenicity, and immunological failure insights.

Microbiology spectrum·2024
Same author

PARP-1 dependent cell death pathway (Parthanatos) mediates early brain injury after subarachnoid hemorrhage.

European journal of pharmacology·2024
Same author

Comparative transcriptome analysis reveals transcriptional regulation of anthocyanin biosynthesis in purple radish (Raphanus sativus L.).

BMC genomics·2024
Same author

The influence of the doping concentration and reverse intersystem crossing on the efficiency of tricomponent organic light-emitting diodes with the thermally activated delayed fluorescence exciplex emitter.

RSC advances·2024
Same author

Quantitative Assessment of Lid Margin Vascularity Using Swept-Source Optical Coherence Tomography Angiography.

Translational vision science & technology·2024

相关实验视频

Updated: Jan 14, 2026

Deep Neural Networks for Image-Based Dietary Assessment
13:19

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

9.9K

一个基于改进的ResNet-34网络的OCT视网膜图像分类模型.

Zhenwei Li1, Jiawen Wang1, Angchao Duan1

  • 1Henan University of Science and Technology, School of Medical Technology and Engineering, Luoyang, 471003, CHINA.

Biomedical physics & engineering express
|October 22, 2025
PubMed
概括
此摘要是机器生成的。

这项研究引入了一种新的CANet模型,用于使用光学连贯断层扫描 (OCT) 图像自动进行视网膜疾病分类. 该模型实现了高精度,有助于早期诊断糖尿病黄斑胀和胆道新血管化等疾病.

关键词:
一个AMP AMP就是一个AMP.这就是为什么CBAM是CBAM.其他国家和地区.在ResNet-34中使用ResNet-34.视网膜疾病分类 视网膜疾病分类

更多相关视频

Efficient and Consistent Generation of Retinal Pigment Epithelium/Choroid Flatmounts from Human Eyes for Histological Analysis
07:59

Efficient and Consistent Generation of Retinal Pigment Epithelium/Choroid Flatmounts from Human Eyes for Histological Analysis

Published on: October 28, 2022

3.3K

相关实验视频

Last Updated: Jan 14, 2026

Deep Neural Networks for Image-Based Dietary Assessment
13:19

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

9.9K
Efficient and Consistent Generation of Retinal Pigment Epithelium/Choroid Flatmounts from Human Eyes for Histological Analysis
07:59

Efficient and Consistent Generation of Retinal Pigment Epithelium/Choroid Flatmounts from Human Eyes for Histological Analysis

Published on: October 28, 2022

3.3K

科学领域:

  • 眼科医生 眼科 眼科
  • 医疗成像医学成像
  • 人工智能的人工智能

背景情况:

  • 视网膜疾病是视力丧失的主要原因,需要早期检测.
  • 从光学一致性断层扫描 (OCT) 图像中手动诊断视网膜疾病是复杂且耗时的.
  • 自动分类方法对于有效和准确的诊断至关重要.

研究的目的:

  • 开发一种自动化分类系统,用于糖尿病黄斑 (DME),冠状腺新血管 (CNV),玻璃状 (Drusen) 和正常的视网膜图像使用OCT.
  • 增强功能提取能力,以提高诊断准确度.
  • 为网膜疾病的早期临床诊断提供可靠的技术解决方案.

主要方法:

  • 设计了一个新的CBAM-AMP网络 (CANet),将卷积块注意模块 (CBAM) 集成到ResNet-34架构中.
  • 采用自动混合精度 (AMP) 和转移学习来加快培训并改善模型通用化.
  • 数据预处理技术包括中位过,规范化和数据增强,用于优化图像质量和解决类不平衡.

主要成果:

  • 在2017年OCT数据集上,CANet模型实现了0.9890的总分类准确性.
  • 曲线下的面积 (AUC) 值在所有类别中达到1,而CNV的回忆值为1.
  • 废弃实验证实了CBAM (0.9%的准确度增加),AMP (1.6%的增加) 和转移学习 (9.4%的增加) 的显著贡献,证明了对杂数据的稳定性.

结论:

  • 集成的CANet模型显著提高了OCT图像分类性能.
  • 拟议的方法为自动视网膜疾病诊断提供了一种高效和强大的解决方案.
  • 这项技术有可能帮助临床医生在早期和准确地识别危及视力的视网膜疾病.