机器学习在卵巢癌诊断与预后中应用的研究进展  

Research Progress on Application of Machine Learning in Diagnosis and Prognosis of Ovarian Cancer

作　　者：王瀚苑 赵之滢 宫婷婷[3] 赵玉虹[1,2] 吴琪俊[1,2] WANG Hanyuan;ZHAO Zhiying;GONG Tingting;ZHAO Yuhong;WU Qijun(Department of Clinical Epidemiology,Shengjing Hospital,China Medical University,Shenyang 110004,China;Clinical Research Center,Shengjing Hospital of China Medical University,Shenyang 110004,China;Department of Obstetrics and Gynecology,Shengjing Hospital,China Medical University,Shenyang 110004,China)

机构地区：[1]中国医科大学附属盛京医院临床流行病学教研室,沈阳110004 [2]中国医科大学附属盛京医院临床研究中心,沈阳110004 [3]中国医科大学附属盛京医院妇产科,沈阳110004

出　　处：《肿瘤防治研究》2021年第8期804-808,共5页Cancer Research on Prevention and Treatment

基　　金：国家重点研发计划(2017YFC0907404);国家自然科学基金面上项目(82073647);中国博士后科学基金面上项目(2018M641752);辽宁省“兴辽英才计划”项目(XLYC1907102);沈阳市中青年科技创新人才支持计划(RC190484);中国医科大学首批健康医疗大数据研究课题(HMB201902107)。

摘　　要：妇科恶性肿瘤之一的卵巢癌对女性健康存在严重威胁。机器学习将统计学和计算机学交叉融合,研究人员将机器学习方法应用于卵巢癌诊断及预后研究,旨在提高临床诊断水平和改善患者预后。本文对机器学习在卵巢癌诊断和预后方面的应用进行综述,结果表明,机器学习模型在卵巢癌诊断和预后方面的预测能力均优于传统统计学模型,但未来需要在前瞻性大规模研究中对各类模型进一步测试和验证。Ovarian cancer,one of the gynecological malignancies,poses a serious threat to women’s health.The machine learning combines statistics and computer science.Researchers apply the machine learning to the clinical diagnosis and prognosis research of ovarian cancer.This article reviews the applications of machine learning in ovarian cancers.The results show that the predictive ability of machine learning models is better than traditional statistical models,but further test and verification are needed in prospective large-scale studies.

关 键 词：卵巢癌 机器学习 诊断 预后 

分 类 号：R737.31[医药卫生—肿瘤]