[1]王景川,胡喜风,许宏吉△,等.全视野数字病理图像智能分析*[J].生物医学工程研究,2024,03:175-180.
WANG Jingchuan,HU Xifeng,XU Hongji,et al.Intelligent analysis of whole slide imaging[J].Journal of Biomedical Engineering Research,2024,03:175-180.
点击复制
《生物医学工程研究》[ISSN:1006-6977/CN:61-1281/TN]
- 期数:
-
2024年03期
- 页码:
-
175-180
- 栏目:
-
- 出版日期:
-
2024-06-25
文章信息/Info
- Title:
-
Intelligent analysis of whole slide imaging
- 文章编号:
-
1672-6278 (2024)03-0175 -06
- 作者:
-
王景川; 胡喜风; 许宏吉△; 刘治△
-
(山东大学 信息科学与工程学院,青岛 266237)
- Author(s):
-
WANG Jingchuan; HU Xifeng; XU Hongji; LIU Zhi
-
(School of Information Science and Engineering,Shandong University,Qingdao 266237,China)
-
- 关键词:
-
深度学习; 全视野数字病理切片; 数字病理学图像分析; 卷积神经网络; 组织病理学图像
- Keywords:
-
Deep learning; Whole slide imaging; Digital pathology image analysis; Convolutional neural networks; Histopathological image
- 分类号:
-
R318
- DOI:
-
10.19529/j.cnki.1672-6278.2024.03.01
- 文献标识码:
-
A
- 摘要:
-
随着数字组织病理学的快速发展,全视野数字病理切片(whole slide imaging,WSI)在医疗领域得到了广泛应用。近年来,深度学习算法的飞速发展为WSI的研究提供了新契机。为更好地分析WSI,充分利用其中丰富的细节信息,通过深度学习算法提取WSI图像特征,进而完成各种下游任务已成为当前的研究热点。本文对WSI图像的智能分析作了综述,首先介绍了利用深度学习进行颜色归一化的方法,随后回顾了不同研究在输入数据筛选方面采用的不同策略。最后,本文总结了深度学习在WSI的分割、分类、预测三大任务中的应用,并探讨了其在WSI应用中面临的挑战和未来的发展方向。
- Abstract:
-
With the rapid advancement of digital histopathology, whole slide imaging (WSI) has seen widespread application in the medical field. In recent years, the rapid development of deep learning algorithms has provided new opportunities for WSI research. To better analyze WSI and fully utilize its rich detailed information, and extract features from WSI images by using deep learning algorithms, thereby accomplishing various downstream tasks has become a research hotpot. We provide a comprehensive review of the intelligent analysis of WSI images. Firstly, several methods for color normalization using deep learning are introduct. Subsequently, we review different strategies employed in various studies for input data selection. Finally, we summarize the applications of deep learning in the three major downstream tasks of WSI images: segmentation, classification, and prediction, and discuss the challenges and future directions for the application of deep learning in WSI.
备注/Memo
- 备注/Memo:
-
(收稿日期:2024-06-03)山东省重点研发计划(重大科技创新工程)(2021CXGC010506);济南高校二十条项目(2021GXRC024)。△通信作者Email:hongjixu@sdu.edu.cn;liuzhi@sdu.edu.cn
更新日期/Last Update:
2024-07-18