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

A meta-analysis on the effectiveness and safety of FOLFOX plus bevacizumab for colorectal cancer treatment.

Frontiers in oncology·2026
Same author

From sparse semantics to rich instances: Empowering label-efficient LiDAR panoptic segmentation via geometric priors.

Neural networks : the official journal of the International Neural Network Society·2026
Same author

A panoramic review of transcranial focused ultrasound neuromodulation: from basic research to clinical applications.

Journal of neuroengineering and rehabilitation·2025
Same author

Therapeutic targeting of STING-IL6/STAT3 axis to inhibit osteoclastic niche formation and breast cancer bone metastasis.

Cell death discovery·2025
Same author

Correlation study of tumor-infiltrating lymphocytes combined with residual cancer burden and prognosis in breast cancer patients receiving neoadjuvant chemotherapy.

Frontiers in oncology·2025
Same author

Isofetamid Sensitivity and SDHI Cross-Resistance in <i>Botrytis cinerea</i> from Strawberry in Shanghai, China.

Plant disease·2025

相关实验视频

Updated: Jan 8, 2026

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

3.3K

加速视觉基础模型,以实现高效的医疗图像细分.

Xian-Tao Wu1, Xiao-Diao Chen1, Wen Wu2

  • 1School of Computer Science and Technology, Hangzhou Dianzi University, Hangzhou, China.

Medical physics
|December 19, 2025
PubMed
概括
此摘要是机器生成的。

这项研究通过降低计算成本来加速医疗图像细分的细分任何模型 (SAM). 一种新的方法将CNN辅助调整与令牌暂停相结合,实现更快的处理和更好的分段质量.

关键词:
深度学习是一种深度学习.医疗图像细分 医疗图像细分细分任何东西模型模型.通过量通过量.标记暂停暂停时间

更多相关视频

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

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

723
Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images
14:08

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

Published on: April 13, 2013

43.4K

相关实验视频

Last Updated: Jan 8, 2026

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

3.3K
Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

723
Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images
14:08

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

Published on: April 13, 2013

43.4K

科学领域:

  • 人工智能的人工智能
  • 医疗成像医学成像
  • 计算机视觉 计算机视觉

背景情况:

  • 分段任何模型 (SAM) 显示出基于人工智能的医疗图像分割的前景.
  • 由于视觉变压器的复杂性,SAM的高计算成本阻碍了实时应用.

研究的目的:

  • 为了加速SAM用于医疗图像细分.
  • 为了提高细分质量和减少内存使用.

主要方法:

  • 提出了一个CNN辅助的调整策略,使SAM能够处理较小的输入,减少补丁和内存.
  • 引入了一个令牌暂停策略,以跳过计算以获得更少信息的补丁,解决冗余性.
  • 结合了这两种策略,实现了高效,可适应的医疗图像细分.

主要成果:

  • 与现有的基于SAM的方法相比,实现了12倍更快的处理速度.
  • 在Synapse和ACDC基准指标上表现出优异的细分性能.
  • 通过处理较小的输入,显著降低了内存消耗.

结论:

  • 确定输入大小调整和统一的补丁处理是医学成像中SAM的局限性.
  • 开发了一个高效的策略,集成基于适配器的调和令牌暂停.
  • 提高了医疗应用的吞吐量和保留了细分性能.