卷积神经网络在前列腺癌诊断中应用的研究进展  

Research progress of convolutional neural networks in the diagnosis of prostate cancer

作　　者：王宇昊[1] 王增军[1] Wang Yuhao;Wang Zengjun(Department of Urology,The First Affiliated Hospital of Nanjing Medical University(Jiangsu Provincial Hospital),Nanjing 210029,China)

机构地区：[1]南京医科大学第一附属医院(江苏省人民医院)泌尿外科,南京210029

出　　处：《中华泌尿外科杂志》2022年第8期637-640,共4页Chinese Journal of Urology

摘　　要：前列腺癌已成为国内男性最高发的泌尿系恶性肿瘤。卷积神经网络是人工智能中最具代表性和发展前景的深度学习算法的代表模型。卷积神经网络在前列腺癌诊断中展示出了优越的性能和应用价值,针对多参数磁共振和病理图像可以达到与影像科和病理科医生相当的诊断效能,并能获得更多的疾病信息;在致病基因的鉴别和基于拉曼光谱的实验室诊断中也有重要作用。本文就卷积神经网络用于前列腺癌影像、病理、基因、实验室等多方面诊断的研究进展进行综述。Prostate cancer has become the one of the most common urinary tract malignant tumor in men in China.Convolutional neural network is one of the representative model and promising deep learning algorithm in artificial intelligence.It has shown superior performance and application value in the diagnosis of prostate cancer.For multi-parameter magnetic resonance and pathological images,it can achieve diagnostic performance equivalent to that of radiologists and pathologists,and could obtain more disease information.There are also important applications of it in identification of disease-causing genes and Raman spectroscopy-based laboratory diagnosis.This article reviews the research progress of convolutional neural networks in the diagnosis of prostate cancer in radiology,pathology,genetics and laboratory.

关 键 词：前列腺肿瘤 卷积神经网络 前列腺癌 诊断 研究进展 

分 类 号：R737.25[医药卫生—肿瘤]