污点变化和颜色规范化对病理学预后预测的影响
Siyu Steven Lin1, Haowen Zhou1, Richard J Cote2
1California Institute of Technology, Department of Electrical Engineering, Pasadena CA 91125 USA.
ArXiv
|September 24, 2024
概括
在病理学中,深度神经网络 (DNN) 与染色变化作斗争. 在一批组织学幻灯片上训练的DNN未能对另一批进行概括,即使在染色正常化后,也突出了对新图像处理方法的需求.
科学领域:
- 数字病理学数字病理学
- 计算病理学计算病理学
- 机器学习在医学中的应用
背景情况:
- 深度神经网络 (DNN) 在病理学方面表现有前途,通过从大型数据集中学习微妙特征,可能超过人类的准确性.
- 组织学幻灯片的染色质的变化对为DNN任务准备数字病理学数据集构成了挑战.
- 污点正常化是一种常见的技术,用于解决数字病理学的图像变异.
研究的目的:
- 评估训练DNN模型对不同时间处理的新批组织学幻灯片的概括性.
- 评估包括CycleGAN在内的染色规范化技术在改善不同批次的DNN模型性能方面的有效性.
- 确定需要新的图像处理和收集策略,以在预测性DNN算法中获得一致的显微镜数据.
主要方法:
- 利用以前报告的DNN模型来识别早期非小细胞肺癌 (NSCLC) 转移.
- 训练并测试了DNN在两个不同的批次的H&E染色原发性瘤组织部分从相同的组织块,在不同的时间处理.
- 应用传统的色调和基于循环生成对抗网络 (CycleGAN) 的染色规范化,以评估它们对跨批量概括的影响.
主要成果:
- 该DNN模型未能对新一批幻灯片进行概括,与同一批表现 (AUC 0.74-0.81) 相比,其预测准确性明显较低 (AUC 0.52-0.53).
- 包括CycleGAN在内的污点规范化方法没有改善DNN的跨批量通用性.
- 幻灯片批次之间的染料变化是限制DNN模型预测性能的关键因素.
结论:
- 在数字病理学中训练有素的DNN模型可能无法对不同时间处理的数据进行概括,即使有染色正常化.
- 目前的染色规范化技术不足以克服幻灯片处理中的时间变化引起的概括差距.
- 开发用于一致的显微镜图像采集和处理的新方法对于可靠和广泛适用的病理学预测DNN算法至关重要.
更多相关视频
22:49Optimized Staining and Proliferation Modeling Methods for Cell Division Monitoring using Cell Tracking Dyes
Published on: December 13, 2012
35.6K
11:27Quantitative Multispectral Analysis Following Fluorescent Tissue Transplant for Visualization of Cell Origins, Types, and Interactions
Published on: September 22, 2013
9.3K
相关概念视频
01:03Fixation and Sectioning
4.2K
Two basic types of preparation are used to visualize specimens with a light microscope: wet mounts and fixed specimens.
The simplest type of preparation is the wet mount, in which the specimen is placed in a drop of liquid on the slide. A liquid specimen can be directly deposited on the slide using a dropper. Solid specimens, such as skin scraping, can be placed on the slide before adding a drop of liquid to prepare the wet mount. Sometimes the liquid is simply water, but stains are often added...
The simplest type of preparation is the wet mount, in which the specimen is placed in a drop of liquid on the slide. A liquid specimen can be directly deposited on the slide using a dropper. Solid specimens, such as skin scraping, can be placed on the slide before adding a drop of liquid to prepare the wet mount. Sometimes the liquid is simply water, but stains are often added...
4.2K
01:17Karyotyping
59.1K
Overview
59.1K