摘要
原发性肝细胞癌(hepatocellular carcinoma,HCC)是最常见的消化系统恶性肿瘤之一,恶性程度高,预后差。微血管侵犯(microvascular invasion,MVI)通常是指在显微镜下于内皮细胞衬覆的血管腔内见到癌细胞巢团。目前MVI普遍被认为与HCC的复发和转移密切相关。因此,术前准确预测MVI具有重要意义。但目前仍缺乏能在术前准确预测MVI的公认有效方法。随着人工智能的兴起与发展,影像组学与深度学习被越来越多用于开发个体化预测模型。影像组学和深度学习技术可以实现影像信息深度挖掘,提供更客观更全面的信息,再结合临床资料建立综合模型,这些模型可以对HCC MVI风险进行准确评估,并帮助医生制订个体化的治疗策略。本文旨在通过对国内外关于影像组学技术对MVI评估的相关研究的综合分析,增强影像医师和临床医师对MVI的认识和关注,并为临床MVI的准确评估与治疗方案的制订,以及HCC患者预后判断提供有益的指导,从而改善HCC患者的诊疗效果,提高生存率,为大数据医疗环境指导下实现个体化精准治疗提供依据。
Hepatocellular carcinoma(HCC)is one of the most common malignant tumors of the digestive system with a high degree of malignancy and poor prognosis.Microvascular invasion(MVI)usually refers to a cluster of cancer cells in the vascular cavity covered by endocrine cells under the microscope.Currently,MVI is generally considered to correlate closely with the recurrence and metastasis of HCC.Therefore,it is essential to predict MVI accurately before surgery.However,there is still no accepted and effective method to accurately predict MVI.With the rise and development of artificial intelligence,radiomics and deep learning are increasingly used to develop individualized predictive models.Radiomics and deep learning technologies can enable deep mining of imaging information to provide more objective and comprehensive information,which can be combined with clinical information to build comprehensive models.These models can accurately assess of HCC MVI risk and help doctors develop individualized treatment strategies.This paper aims to make a comprehensive analysis of relevant studies on MVI assessment by radiomics techniques at home and abroad to enhance the understanding and attention of radiologists and clinicians on MVI and to provide helpful guidance for the accurate assessment and treatment planning of clinical MVI,as well as the prognosis judgment of HCC patients,to improve the diagnosis and treatment results of HCC patients,increase the survival rate,and provide a basis for the realization of the big data medical environment guided by this will provide a basis for individualized and precise treatment under the guidance of big data medical environment.
作者
刘阳
姜艳丽
樊凤仙
杨文霞
李大瑞
刘光耀
张静
LIU Yang;JIANG Yanli;FAN Fengxian;YANG Wenxia;LI Darui;LIU Guangyao;ZHANG Jing(Department of Magnetic Resonance,Lanzhou University Second Hospital,Lanzhou 730030,China;Second Clinical School,Lanzhou University,Lanzhou 730030,China;Gansu Province Clinical Research Center for Functional and Molecular Imaging,Lanzhou 730030,China)
出处
《磁共振成像》
CAS
CSCD
北大核心
2023年第9期159-164,共6页
Chinese Journal of Magnetic Resonance Imaging
基金
甘肃省科技计划项目(编号:21JR11RA122、21JR7RA438)。
关键词
肝细胞癌
微血管侵犯
影像组学
术前预测
深度学习
hepatocellular carcinoma
microvascular invasion
radiomics
preoperative prediction
deep learning
