面向前列腺癌Gleason评分的深度学习方法研究现状综述  

A Review of the Current State of Research on Deep Learning Methods for Gleason Scoring of Prostate Cancer

作　　者：李嘉 蔡定建 黄丽 杨敏 Li Jia;Cai Dingjian;Huang Li;Yang Min(Faculty of Intelligent Manufacturing,Wuyi University,Jiangmen,Guangdong 529020,China)

机构地区：[1]五邑大学智能制造学部,广东江门529020

出　　处：《机电工程技术》2024年第2期56-59,93,共5页Mechanical & Electrical Engineering Technology

基　　金：江门市科技计划资助项目(2022JC01023)。

摘　　要：前列腺癌是一种常见的男性疾病,严重危害着男性的身体健康。前列腺癌Gleason评分对于诊断前列腺癌及判断患者治疗后的情况,具有非常重要的意义。随着深度学习技术的快速发展,涌现出了大量基于深度学习的前列腺癌Gleason评分方法。以不同类型医学临床数据为研究对象,可将现有面向前列腺癌Gleason评分的深度学习方法归纳为面向核磁共振图像、面向活检组织病理图的前列腺癌Gleason评分方法。围绕上述研究对象,对现有面向前列腺癌Gleason评分的深度学习方法进行分析和总结,并对相关技术的发展趋势进行展望。Prostate cancer is a common male disease that seriously endangers men’s health.Gleason score for prostate cancer is very important for diagnosing prostate cancer and judging patients’condition after treatment.With the rapid development of deep learning technology,a large number of deep learning-based Gleason scoring methods for prostate cancer have emerged.Taking different types of medical clinical data as the research objects,the existing deep learning methods for Gleason scoring of prostate cancer can be summarized into Gleason scoring methods for magnetic resonance images and Gleason scoring methods for biopsy histopathology images of prostate cancer.The existing deep learning methods for Gleason score of prostate cancer are analyzed and summarized,and the development trend of related technologies is prospected.

关 键 词：前列腺癌 GLEASON评分 深度学习 核磁共振图像 活检组织病理图 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]