HCTTI:用于组织图像污点规范化的高性能异质计算工具包
Yan Jiang1,2,3, Bo Wang4, Weipeng Xing5
1School of Software, Quanzhou University of Information Engineering, Quanzhou, Fujian, 362000, China. jianghnu@hnu.edu.cn.
Journal of imaging informatics in medicine
|January 17, 2025
概括
一个新的工具包,HCTTI,显著加速整个幻灯片成像 (WSI) 分析,用于癌症诊断的深度学习. 它优化了WSI读取,规范化和保存,使得医疗诊断更快,更有效.
科学领域:
- 数字病理学数字病理学
- 计算医学是一种计算医学.
- 医疗图像分析 医疗图像分析
背景情况:
- 整体幻灯片成像 (WSI) 对于癌症诊断和治疗至关重要,通过深度学习实现先进的医学诊断.
- 在不同机构的WSI中,颜色和强度的变化阻碍了深度学习分类的准确性.
- 目前的WSI处理和规范化在很大程度上依赖于CPU,从而导致低于最佳的计算性能.
研究的目的:
- 引入用于组织图像的高性能异构计算工具包 (HCTTI),旨在优化WSI分析.
- 为了评估各种WSI阅读器,颜色规范化技术和序列化格式的性能.
- 为了证明HCTTI在分布式和多节点GPU环境中的效率提升.
主要方法:
- 开发了HCTTI,集成了系统级优化,用于WSI读取,规范化和节省.
- 对比HCTTI的性能与现有的工具,如OpenSlide和TIAToolbox用于WSI读取和规范化.
- 评估不同序列格式 (HDF5,PNG,Zarr) 的效率,以存储正常化组织图像.
- 实现多节点分布式GPU处理以提高可扩展性.
主要成果:
- 与OpenSlide相比,HCTI在WSI读数中实现了7倍的加快速度.
- 在HCTTI中,GPU加速的Macenko正常化比TIAToolbox实现速度快9倍.
- HDF5在存储正常化图像方面表现出卓越的性能,提供13倍更快的写入速度和2倍更快的阅读速度.
- 与TIAToolbox.com相比,使用HCTI对单个WSI进行正常化时观察到13倍的加快速度.
- 通过多节点分布式GPU实现实现了线性加速.
结论:
- HCTTI为分布式WSI读取,规范化和序列化提供了全面的解决方案,显著提高了处理效率.
- 该工具包的性能增强有潜力加速基于深度学习的WSI分析,用于医学诊断.
- HCTTI的优化可以为更有效和高效的癌症诊断和治疗策略做出贡献.
相关概念视频
Computed Tomography
7.6K
Tomography refers to imaging by sections. Computed tomography (CT) is a non-invasive imaging technique that uses computers to analyze several cross-sectional X-rays to reveal minute details about structures in the body.
The technique was invented in the 1970s and is based on the principle that as X-rays pass through the body, they are absorbed or reflected at different levels. In the technique, a patient lies on a motorized platform while a computerized axial tomography (CAT) scanner rotates...
The technique was invented in the 1970s and is based on the principle that as X-rays pass through the body, they are absorbed or reflected at different levels. In the technique, a patient lies on a motorized platform while a computerized axial tomography (CAT) scanner rotates...
7.6K
Statgraphics
485
Statgraphics is a comprehensive statistical software suite designed for both basic and advanced data analysis. Originating in 1980 at Princeton University under Dr. Neil W. Polhemus, it was one of the pioneering tools for statistical computing on personal computers, with its public release in 1982 marking an early milestone in data science software. Over the years, it has evolved into a robust platform for data science, offering tools for regression analysis, ANOVA, multivariate statistics,...
485


