检索规则说明:AND代表“并且”;OR代表“或者”;NOT代表“不包含”;(注意必须大写,运算符两边需空一格)
检 索 范 例 :范例一: (K=图书馆学 OR K=情报学) AND A=范并思 范例二:J=计算机应用与软件 AND (U=C++ OR U=Basic) NOT M=Visual
作 者:李文华 许传斌[1] 崔景景 刘晓鸣 王肖 贾文昊 李天然[2] LI Wen-hua;XU Chuan-bin;CUI Jing-jing;LIU Xiao-ming;WANG Xiao;JIA Wen-hao;LI Tian-ran(The Clinical Medical College of Jiamusi University,Jiamusi 154002,Heilongjiang,China;Department of Radiology,the Fourth Medical Center of PLA General Hospital,Beijing 100048,China;United Imaging Intelligence
机构地区:[1]佳木斯大学临床医学院,黑龙江佳木斯154002 [2]解放军总医院第四医学中心放射诊断科,北京100048 [3]联影医疗科技<北京>有限公司,北京100089
出 处:《医学信息》2022年第18期39-43,52,共6页Journal of Medical Information
基 金:解放军总医院临床科研扶持基金(编号:2017FC-304Z-GLCX-01)。
摘 要:目的初步探讨基于治疗前T_(2)加权成像(T2WI)的影像组学特征预测肝细胞癌(HCC)射频消融术(RFA)后早期复发的价值。方法回顾性分析2015年1月-2021年3月于解放军总医院第四医学中心采取射频消融治疗的140例肝细胞癌患者的临床和影像资料,术后定期随访,了解是否发生早期复发。所有样本按8∶2的比例随机分成训练集和测试集,在训练集中先提取734个特征,采用方差阈值、K最佳和LASSO算法筛选特征,最后构建机器学习预测模型(逻辑回归、支持向量机、随机森林)并采用测试集进行内部验证。结果共筛选出8个与肝细胞癌早期复发相关的特征,随机森林分类器在训练集预测肝细胞癌术后早期复发的AUC值、准确率、诊断敏感性、特异性分别为0.903、0.812、0.839、0.802;在测试集中的AUC值、准确率、诊断敏感性、特异度分别为0.806、0.750、0.625、0.800。结论治疗前T2WI的影像组学特征可用于肝细胞癌射频消融术后早期复发的自动化评估,这有助于临床医生制定合适治疗方案。Objective To preliminarily explore the diagnostic value of radiomics features frompreoperative T_(2)-weighted imaging(T2WI)in predicting the early recurrence of hepatocellular carcinoma after radiofrequency ablation.Methods Retrospectively analyze the clinical and imaging data of 140 hepatocellular carcinoma patients with radiofrequency ablation treatment at the Fourth Medical Center of PLA General Hospital from January 2015 to March 2021.Early recurrence was confirmed by postoperative regular follow-up.All samples are randomly divided into training set and validation set with a percentage of 8:2.In the training set,734 features were extracted first,and the features were screened by variance threshold,K best and LASSO algorithm.Finally,the machine learning prediction model(logistic regression,support vector machine,random forest)was constructed and the test set was used for internal verification.Results A total of 8 features related to early recurrence of hepatocellular carcinoma were screened out.The AUC value,accuracy,diagnostic sensitivity and specificity of random forest classifier in predicting early recurrence of hepatocellular carcinoma after operation in training set were 0.903,0.812,0.839 and 0.802,respectively.The AUC value,accuracy,diagnostic sensitivity and specificity in the test set were 0.806,0.750,0.625 and 0.800,respectively.Conclusion The radiomics features of T2WI before treatment can be used for automatic evaluation of early recurrence after radiofrequency ablation of hepatocellular carcinoma,which is helpful for clinicians to formulate appropriate treatment plans.
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在链接到云南高校图书馆文献保障联盟下载...
云南高校图书馆联盟文献共享服务平台 版权所有©
您的IP:216.73.216.171