Search research articles

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 Journal Methods Collections JoVE Encyclopedia of Experiments 存档

教育

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual 教师资源中心教师网站

使用条款与条件

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

[Prevalence and prognostic factors for postoperative complications of uvulopalatopharyngoplasty in patients with obstructive sleep apnea hypopnea syndrome].

Lin chuang er bi yan hou tou jing wai ke za zhi = Journal of clinical otorhinolaryngology head and neck surgery·2008

Same author

[Transurethral electrotomy for cystis vesicular seminalis induced by obstruction of the distal end of the ejaculatory duct].

Zhonghua nan ke xue = National journal of andrology·2008

Same author

[Effects of testosterone on the proliferation of rat corpus cavernosum cells in vitro].

Zhonghua nan ke xue = National journal of andrology·2008

Same author

Identification of 4-aminopyrazolylpyrimidines as potent inhibitors of Trk kinases.

Journal of medicinal chemistry·2008

Same author

Increased dialysate levels of phospholipids containing unsaturated fatty acid are associated with increased peritoneal transport rate.

American journal of nephrology·2008

Same author

Stepwise increase in the prevalence of isolated systolic hypertension with the stages of chronic kidney disease.

Nephrology, dialysis, transplantation : official publication of the European Dialysis and Transplant Association - European Renal Association·2008

Search research articles

相关实验视频

Updated: Jul 21, 2025

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

半监督医疗图像细分与联合分配对齐.

Tao Wang^1,2,3, Zhongzheng Huang², Jiawei Wu⁴

¹Fujian Provincial Key Laboratory of Information Processing and Intelligent Control, College of Computer and Control Engineering, Minjiang University, Fuzhou 350108, China.

Bioengineering (Basel, Switzerland)

|July 29, 2023

概括

本研究引入了用于半监督医疗图像细分的共同分布对齐 (Co-DA),有效地解决了数据稀缺性和阶级不平衡问题. 通过对模型之间的预测进行对齐,Co-DA提高了对少数群体类别的细分精度,优于现有方法.

关键词:

联合培训是指联合培训.分布线对齐的调整情况.医疗图像细分医疗图像细分半监督学习半监督学习

更多相关视频

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

Published on: April 13, 2013

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

相关实验视频

Last Updated: Jul 21, 2025

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

Published on: April 13, 2013

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

科学领域:

医疗成像医学成像
计算机视觉计算机视觉
机器学习机器学习

背景情况:

医疗图像细分在很大程度上依赖于大型标记数据集,由于需要专家注释,这些数据集的创建是昂贵和耗时的.
医疗图像中的类别分布不均,往往导致代表性不足的类别的表现不佳,阻碍诊断准确性.
当前的半监督方法在医疗图像细分中面临着有限的标记数据和不平衡类的挑战.

研究的目的:

提出一种新的半监督方法,即联合分布对齐 (Co-DA),用于医疗图像细分.
解决医疗图像细分数据集中昂贵的注释和类不平衡的局限性.
为了提高细分性能,特别是少数群体,使用有限的标记数据.

主要方法:

同分配对齐 (Co-DA) 在未标记的数据上对准两个不同初始化的模型之间的类智能边际预测.
由一个模型生成的伪标签用于监督另一个模型,促进知识传输.
引入了一个过度预期的交叉损失,以从未标记的像素中过噪音的伪标签.

主要成果:

与最先进的方法相比,Co-DA在公共2D CaDIS,3D LGE-MRI和ACDC数据集上取得了更高的性能.
该方法在CaDIS上使用仅24%的标记数据获得了0.8515的mIoU.
在LGE-MRI上的0.8824和ACDC上的0.8773的子得分仅在20%的标记数据中实现.

结论:

联合分发对齐 (Co-DA) 是医疗图像细分的有效半监督策略,显著减少对广泛注释的依赖.
提出的方法即使在有限的标记数据下也显示出强大的性能,并解决了阶级不平衡的挑战.
在资源有限的场景中,Co-DA为提高医疗图像细分效率和准确性提供了一个有希望的解决方案.