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:魏明翔 丁聪 严彩英 柏根基[3] 陈双庆[1,2] WEI Mingxiang;DING Cong;YAN Caiying(Department of Radiology,The Affiliated Suzhou Hospital of Nanjing Medical University,Suzhou Municipal Hospital Nanjing,Jiangsu Province 215001,P.R.China)
机构地区:[1]南京医科大学附属苏州医院放射科苏州市立医院,苏州215001 [2]南京医科大学姑苏学院,苏州211166 [3]南京医科大学附属淮安第一医院影像中心,淮安223300
出 处:《临床放射学杂志》2023年第10期1636-1641,共6页Journal of Clinical Radiology
摘 要:目的探究基于多参数MRI影像组学和机器学习方法在术前鉴别诊断Ⅰ型和Ⅱ型上皮性卵巢癌(EOC)中的价值。方法回顾性搜集两个中心共181例EOC患者(中心一136例为训练集,中心二45例为外部验证集),其中Ⅰ型59例,Ⅱ型122例。从每例患者抑脂(FS)-T2WI、DWI及ADC图像分别提取1130个影像组学特征。通过对四种机器学习算法的性能评价,确定了构建影像组学模型的理想算法。构建影像组学模型、临床模型和联合模型,并通过受试者工作特征(ROC)曲线分析评估诊断性能。采用DeLong检验比较曲线下面积(AUC)。结果随机森林(RF)算法是构建影像组学模型的最优算法。联合模型在外部验证集AUC为0.912(95%CI:0.820~1.000),显著优于临床模型(AUC=0.718,95%CI:0.552~0.884,P=0.036)和影像组学模型(AUC=0.810,95%CI:0.675~0.946,P=0.012)。结论基于多参数MRI的影像组学和机器学习方法有潜力术前准确鉴别Ⅰ型和Ⅱ型EOC,并协助临床决策。Objective To evaluate the performance of multiparametric MRI-based radiomics and machine learning models in preoperatively discriminating typeⅠand typeⅡepithelial ovarian cancer(EOC).Methods A total of 181 patients with EOC were retrospectively included from two different centers(136 cases in center 1 were used as the training set,and 45 cases in center 2 were used as the external validation set),59 with typeⅠand 122 with typeⅡ.A total of 1130 radiomics features were extracted from each patient's FS-T2WI,DWI,and ADC images,respectively.The ideal algorithm for building the radiomics model was determined by evaluating the performance of four machine learning algorithms.The radiomics model,the clinical model,and the combined model were constructed,and the diagnostic performance was assessed by ROC analysis.The DeLong test was used to compare the area under the ROC curves.Results Random forest was the optimal algorithm for constructing the radiomics model.The combined model had an AUC of 0.912(95%CI:0.820-1.000),which outperformed either the clinical model(AUC=0.718,95%CI:0.552-0.884,P=0.036)or the radiomics model(AUC=0.810,95%CI:0.675-0.946,P=0.012)alone in the external validation set.Conclusion The multiparameter MRI-based radiomics machine learning approach has the potential to preoperatively discriminate typeⅠand typeⅡEOC and assist in clinical decision-making.
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