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

Augmented Reality Navigation-Assisted Laparoscopic Hepatectomy: A Retrospective Comparison of Anatomical and Non-anatomical Resection.

Clinical Medicine Insights. Oncology·2026
Same author

Correspondence-free local-to-global liver deformation correction via implicit neural representation and biomechanical model.

International journal of computer assisted radiology and surgery·2026
Same author

Liver point cloud registration via multi-feature fusion and hyperbolic embedding for augmented reality surgical navigation.

Computer methods and programs in biomedicine·2026
Same author

Laparoscopic liver point cloud registration via hyperbolic-topology interaction.

International journal of computer assisted radiology and surgery·2026
Same author

Digs: diffusion-guided Gaussian Splatting for dynamic occlusion surgical scene reconstruction.

International journal of computer assisted radiology and surgery·2026
Same author

A Landmark-Free 3D-2D Rigid Liver Registration via Point Cloud Matching for Laparoscopic Surgery.

Healthcare technology letters·2025

相关实验视频

Updated: Jun 30, 2025

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

396

SwinD-Net:用于腹腔镜肝脏细分的轻量级细分网络.

Shuiming Ouyang1,2, Baochun He1,2, Huoling Luo1

  • 1Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences, Shenzhen, China.

Computer assisted surgery (Abingdon, England)
|March 20, 2024
PubMed
概括

我们开发了SwinD-Net,这是一种轻量级的深度学习模型,用于实时手术图像细分. 它实现了高精度,显著降低了计算成本,使其适合缺乏强大资源的医院.

关键词:
图像细分 图像细分 图像细分深度学习是一种深度学习.腹腔镜肝脏外科手术 腹腔镜肝脏手术轻量级的模型轻量级的模型.

更多相关视频

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

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

Published on: November 30, 2022

2.7K
Laparoscopic Anatomical Liver Segment VII Resection with Liver Parenchymal Transection Following a Priority Approach
13:57

Laparoscopic Anatomical Liver Segment VII Resection with Liver Parenchymal Transection Following a Priority Approach

Published on: May 23, 2025

229

相关实验视频

Last Updated: Jun 30, 2025

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

396
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

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

Published on: November 30, 2022

2.7K
Laparoscopic Anatomical Liver Segment VII Resection with Liver Parenchymal Transection Following a Priority Approach
13:57

Laparoscopic Anatomical Liver Segment VII Resection with Liver Parenchymal Transection Following a Priority Approach

Published on: May 23, 2025

229

科学领域:

  • 医学图像分析 医学图像分析
  • 计算机视觉 计算机视觉
  • 手术技术 手术技术

背景情况:

  • 实时图像细分对于腹腔镜外科辅助系统至关重要.
  • 传统的深度学习模型提供了高精度,但计算密集,限制了它们在资源有限的医院环境中的使用.
  • 需要有效的细分模型,平衡准确性和计算开销.

研究的目的:

  • 提出一种新的,轻量级的深度学习网络SwinD-Net,用于腹腔镜外科手术中的实时图像细分.
  • 为了减少计算负担和参数数量,同时保持高细分精度.
  • 在CholecSeg8k数据集上验证SwinD-Net的有效性.

主要方法:

  • 开发了包含Skip连接,深度可分离卷积和Swin变压器块的SwinD-Net.
  • 通过消除第一层跳过连接和减少浅功能地图通道来优化网络.
  • 引入了Swin变压器块来捕获全球信息和高级语义特征.

主要成果:

  • 在CholecSeg8k数据集上,SwinD-Net实现了高精度,并大大降低了计算开销.
  • 该模型只需要98.82M FLOP和0.52M参数,在CPU上推断时间为47.49ms/image.
  • 在子度量方面表现优于UNeXt,参数为1/3,FLOP为1/22,推断速度快2.4倍.

结论:

  • SwinD-Net有效地减少了参数数量和计算复杂性,提高了推断速度,同时保持了可比的准确性.
  • 轻量级的设计使得SwinD-Net适合在医院实时应用,而医院的计算资源有限.
  • 拟议的网络为手术图像细分的准确性,速度和效率提供了全面的改进.