相关实验视频

Updated: May 5, 2026

Exploiting Live Imaging to Track Nuclei During Myoblast Differentiation and Fusion

Exploiting Live Imaging to Track Nuclei During Myoblast Differentiation and Fusion

Published on: April 13, 2019

对于注释效率高的核心实例细分的少数镜头学习

Yu Ming, Zihao Wu, Jie Yang

IEEE transactions on medical imaging

|April 25, 2025

概括

核实例细分在组织病理学图像中现在使用一种新的几次拍摄学习 (FSL) 方法更有效. 这种方法显著减少了注释需求,只用10%的数据实现了接近完全监督方法的性能.

科学领域:

计算病理学计算病理学
医疗图像分析医学图像分析
深度学习是一种深度学习.

背景情况:

核实例细分在组织病理学中至关重要,但需要费力,专家依赖的注释.
像弱监督/半监督学习这样的注释效率高的深度学习方法正在引起人们的兴趣.
利用现有的全注释数据集可以帮助对有限注释目标数据集进行细分.

研究的目的:

开发一种使用少数镜头学习 (FSL) 进行核心实例细分的注释效率高的方法.
通过将FSL扩展到通用少数镜头实例细分 (GFSIS) 来适应核心细分.
纳入结构性指导,以应对诸如接触细胞和异质性等挑战.

主要方法:

提出了一个结构指导的通用化几击实例细分 (SGFSIS) 框架.
向GFSIS扩展了少数镜头实例分割 (FSIS),以处理数据集之间的不一致类.
整合了一个结构指导机制,以提高对具有挑战性的核特征的细分精度.

主要成果:

与其他注释效率高的方法 (半监督,转移学习) 相比,SGFSIS表现优越.
实现了与完全监督学习相比的性能,仅使用大约10%的注释.
在多个公开可访问的组织病理学数据集中验证了有效性.

结论:

SGFSIS框架为注释效率高的核心实例细分提供了一个强大的解决方案.
这种方法显著减少了计算病理学中的注释负担.
该方法在需要高精度细分与最小标记数据的实际应用中表现有希望.

更多相关视频

Mapping the Emergent Spatial Organization of Mammalian Cells using Micropatterns and Quantitative Imaging

Mapping the Emergent Spatial Organization of Mammalian Cells using Micropatterns and Quantitative Imaging

Published on: April 30, 2019

Low-input Nucleus Isolation and Multiplexing with Barcoded Antibodies of Mouse Sympathetic Ganglia for Single-nucleus RNA Sequencing

Low-input Nucleus Isolation and Multiplexing with Barcoded Antibodies of Mouse Sympathetic Ganglia for Single-nucleus RNA Sequencing

Published on: March 23, 2022

相关概念视频

The Nucleus

The Nucleus

The nucleus is a membrane-bound organelle that acts as a control center in a eukaryotic cell. It contains chromosomal DNA, which controls gene expression and precisely regulates the production of proteins within the cell. In contrast, the DNA inside the mitochondria and chloroplast only carries out functions that are specific to those organelles.
Arrangement of DNA within Nucleus
The regulation of gene expression inside the nucleus is dependent on many factors, including the DNA structure. The...

Insensitive Nuclei Enhanced by Polarization Transfer (INEPT)

Insensitive Nuclei Enhanced by Polarization Transfer (INEPT)

Insensitive Nuclei Enhanced by Polarization Transfer (INEPT) is an advanced Nuclear Magnetic Resonance (NMR) technique specifically designed to detect and enhance the signals of low-abundance nuclei, such as carbon-13 and nitrogen-15, in small molecules. The fundamental principle behind INEPT is the transfer of polarization from a more abundant and highly polarizable nucleus, typically hydrogen-1, to the low-abundance nucleus of interest. This process effectively boosts the NMR signal of the...