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

Synthesis of starch-based non-isocyanate polyurethane hybrids with hot-melt processability and enhanced performance.

International journal of biological macromolecules·2026

Same author

Efficient and Clean Method for Nano-Lignocellulose Powder by Particle Collision With Magnetic Acceleration.

Small methods·2025

Same author

Toward Next-Generation Semiartificial Photosynthesis: Multidisciplinary Engineering of Biohybrid Systems.

Chemical reviews·2025

Same author

Biohybrid-based pyroelectric bio-denitrification driven by temperature fluctuations.

Nature communications·2025

Same author

Asymmetric Adaptive Heterogeneous Network for Multi-Modality Medical Image Segmentation.

IEEE transactions on medical imaging·2025

Same author

Unlocking interfacial electron transfer in biophotoelectrochemical processes: Role of extracellular polymeric substances in aquatic environments.

Water research·2025

Search research articles

相关实验视频

Updated: Jul 1, 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

多尺度对抗性学习与困难区域监督学习模型用于初级瘤细分的模型.

Shenhai Zheng^1,2, Qiuyu Sun¹, Xin Ye¹

¹College of Computer Science and Technology, Chongqing University of Posts and Telecommunications, Chongqing, People's Republic of China.

Physics in medicine and biology

|March 12, 2024

概括

这项研究引入了一种新的级联深度学习方法,用于精确的瘤细分,结合了多规模的对抗性学习和难以区域监督. 这种方法提高了准确性,特别是在诸如小瘤等具有挑战性的病例中.

关键词:

具有对抗性的学习.一个布式的布式.监管很困难,监管很困难.瘤细分瘤的细分

更多相关视频

Volume Segmentation and Analysis of Biological Materials Using SuRVoS Super-region Volume Segmentation Workbench

Volume Segmentation and Analysis of Biological Materials Using SuRVoS Super-region Volume Segmentation Workbench

Published on: August 23, 2017

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 1, 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

Volume Segmentation and Analysis of Biological Materials Using SuRVoS Super-region Volume Segmentation Workbench

Volume Segmentation and Analysis of Biological Materials Using SuRVoS Super-region Volume Segmentation Workbench

Published on: August 23, 2017

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
人工智能的人工智能
计算机视觉计算机视觉

背景情况:

深度学习在自动瘤细分方面表现出色,但由于瘤的不同特征和不可预测的空间分布,仍然存在挑战.
现有的方法与复杂的瘤形状,类型和不可预测的空间安排作斗争,需要先进的细分技术.

研究的目的:

开发基于级联的深度学习方法,结合多层次的对抗性学习和困难区域监督,以改善瘤细分.
通过提高准确性和处理复杂病例来解决当前瘤细分技术的局限性.

主要方法:

一个粗细的策略,使用多阶段级级二进制细分来简化复杂的细分任务.
引入一个多层次的对抗性学习难度监督UNet (MSALDS-UNet) 具有多个区分器,以提高细分精度.
实施艰难的区域监督损失,以利用结构信息来分割模糊的区域,例如模糊的边界.

主要成果:

拟议的MSALDS-UNet模型在三个独立的公共数据集 (KiTS21,MSD的大脑和胰腺) 上取得了令人满意的结果.
使用关键指标评估绩效:子相似系数,雅卡德相似系数和HD95,证明有效性.
级联方法显著改善了细分性能,特别是在小型瘤对象中.

结论:

开发的级联方法结合了多层次的对抗性学习和困难的监督,为瘤细分提供了精确的解决方案.
这种方法在增强细分性能方面被证明是有效的,特别是对于具有挑战性的小物体,推进自动化医疗图像分析.
这项研究证实了整合多规模对抗性学习和难以监督的协同效益,以实现强大的瘤细分.