基于机器学习的影像组学技术在泌尿系统肿瘤中的临床研究进展  

Advances in clinical research in urologic neoplasms with machine learning-based radiomics technology

作　　者：张云峰 王超[2] 乔小妮[3] 王梦雨 周逢海 ZHANG Yunfeng;WANG Chao;QIAO Xiaoni;WANG Mengyu;ZHOU Fenghai(The First Clinical Medical College of Gansu University of Chinese Medicine,Lanzhou 730000,China;Department of Urology,Gansu Provincial People's Hospital,Lanzhou 730000,China;Department of Information Management,Gansu Provincial People's Hospital,Lanzhou 730000,China;School of Information Science and Engineering,Lanzhou University,Lanzhou 730000,China)

机构地区：[1]甘肃中医药大学第一临床医学院,兰州730000 [2]甘肃省人民医院泌尿外科,兰州730000 [3]甘肃省人民医院信息管理处,兰州730000 [4]兰州大学信息科学与工程学院,兰州730000

出　　处：《磁共振成像》2023年第2期197-202,共6页Chinese Journal of Magnetic Resonance Imaging

基　　金：甘肃省重点研发计划资助项目(编号:21YF5FA016)。

摘　　要：近年来泌尿系统肿瘤的发病率逐年增高,肾癌、膀胱癌(bladder cancer,BCa)、前列腺癌(prostate cancer,PCa)等肿瘤已经成为威胁中老年人健康的重要因素,对泌尿系统恶性肿瘤的早期发现以及预后监测日益成为当前研究的热点。影像组学作为近年新兴起的诊断手段,它通过对组织异质性特征的提取分析,可以无创、定量地对组织进行评价,与传统的影像学检查相比,能更准确地对病灶进行诊断及鉴别诊断。本文从泌尿外科临床医生的角度,就当前影像组学在泌尿系统肿瘤的术前诊断、疗效评价、预后评估、基因表达等方面的研究进展进行综述,并对未来研究方向进行展望。In recent years,the incidence of tumors of the urinary system has increased year by year,kidney cancer,bladder cancer(BCa),prostate cancer(PCa)have become important factors threatening the health of middle-aged and elderly people.The early detection and prognosis monitoring of urinary malignant tumors have increasingly become the hot spot of current research.Radiomics is an emerging diagnostic method in recent years,it enables non-invasive and quantitative evaluation of tissues by extracting and analyzing the characteristics of tissue heterogeneity,compared with traditional imaging,it can diagnose and differentiate lesions more accurately.From urology clinician’s perspective,this paper reviews the current research progress of radiomics in preoperative diagnosis,efficacy evaluation,prognosis evaluation,and gene expression of urologic tumors.

关 键 词：肾上腺肿瘤 肾癌 膀胱癌 前列腺癌 影像组学 磁共振成像 

分 类 号：R445.2[医药卫生—影像医学与核医学] R737[医药卫生—诊断学]