应用肝影像报告和数据系统MRI分级标准诊断原发性肝癌的诊断效能评价

发布时间：2018-05-04 10:11

本文选题：肝影像报告和数据系统 + 磁共振成像　；参考：《中南大学学报(医学版)》2016年04期

【摘要】：目的:初步探讨肝影像报告和数据系统(liver imaging reporting and data system,LI-RADS)磁共振成像(magnet ic resonance imag ing,MR I)分级标准对原发性肝癌的诊断价值。方法:搜集2014年1月至2015年1月在中南大学湘雅医院术后病理证实的100例(原发性肝癌72例,再生结节4例,血管瘤13例,肝囊肿4例,肝脓肿3例,炎性假瘤2例,肝动静脉瘘1例,局限性脂肪变性1例)具有原发性肝癌高危因素患者的MRI影像资料,由3位影像学副教授在不知道病理结果的前提下,严格按照LI-RADS MRI分级标准对图像进行评阅,计算出LI-RADS MRI分级标准诊断原发性肝癌的敏感性、特异性及ROC曲线下面积。结果:应用LI-R ADS MRI分级标准诊断100例患者为原发性肝癌的ROC曲线下面积为0.925。100例患者图像的肝病变分类中LR1类和LR2类20例,术后病理检查均为良性(阴性预测值为100%),31例LR3类中术后病理检查有25例为原发性肝癌,29例LR4类中术后病理检查有27例为原发性肝癌(阳性预测值为93.1%),20例LR5类中术后病理检查表明全部为原发性肝癌(阳性预测值为100%)。将LR3,4,5类归为阳性,则该分级标准的敏感性和特异性分别为100%和71.4%,ROC曲线下面积为0.925。若将LR4,5类归为阳性,则敏感性和特异性可以达到100%和90.9%,ROC曲线下面积为0.974。结论:采用LI-RADS的MRI分级标准诊断原发性肝癌有很好的敏感性及特异性,具有较高的诊断正确率。
[Abstract]:Objective: to evaluate the diagnostic value of imaging imaging reporting and data system (LI-RADS) magnetic resonance imaging (MRI) and magnetic resonance imaging (MRI) classification for primary liver cancer (HCC). Methods: from January 2014 to January 2015, 100 cases (72 cases of primary liver cancer, 4 cases of regenerative nodules, 13 cases of hemangioma, 4 cases of hepatic cyst, 3 cases of liver abscess and 2 cases of inflammatory pseudotumor) were collected from Xiangya Hospital of Central South University. 1 case of hepatic arteriovenous fistula and 1 case of localized fatty degeneration) with MRI imaging data of patients with high risk factors of primary liver cancer. The sensitivity, specificity and area under the ROC curve of the LI-RADS MRI grading criteria for the diagnosis of primary liver cancer were calculated according to the LI-RADS MRI classification criteria. Results: according to the LI-R ADS MRI classification standard, 20 cases of LR1 and LR2 were classified under the ROC curve and the area under the ROC curve was 0.925.100 cases. All postoperative pathological examinations were benign (negative predictive value was 100%) in 31 cases of LR3, of which 25 cases were primary hepatic carcinoma (LR4), 27 cases were primary liver cancer (positive predictive value was 93.1%) in 20 cases of LR5. Post-pathological examination showed that all patients were primary liver cancer (the positive predictive value was 100%). The sensitivity and specificity of the classification criteria were 100% and 71.4%, respectively, and the area under the ROC curve was 0.925. The sensitivity and specificity can reach 100% and 90.9% respectively, and the area under the ROC curve is 0.974 if the LR4N5 is classified as positive. Conclusion: the LI-RADS MRI classification standard has good sensitivity and specificity in the diagnosis of primary liver cancer, and has a high diagnostic accuracy.
【作者单位】：中南大学湘雅医院放射科;
【基金】：湖南省自然科学基金(13JJ6007) 湖南省教育厅项目(13CS13038)~~
【分类号】：R445.2;R735.7

【相似文献】