论文部分内容阅读
目的:评估Tn 2*定量成像（Tn 2* mapping）和基于并行发射平台选择性激发成像的体素内不相干运动（ZOOMit IVIM）序列对甲状腺良恶性结节的诊断效能。n 方法:回顾性收集2019年7月至2020年3月马鞍山市人民医院经手术病理证实的67例甲状腺结节患者资料，共80个结节，其中良性结节39个，恶性结节41个。所有患者术前均行MRI Tn 2WI、Tn 2* mapping及ZOOMit IVIM序列扫描，测量结节的Tn 2*及纯扩散系数（D）、伪扩散系数（Dn *）、灌注分数（f）值。良恶性结节间各参数的比较采用独立样本n t检验或Wilcoxon秩和检验。通过二分类logistic回归建立联合参数模型并计算预测概率值。采用ROC曲线评估各参数和联合参数模型对甲状腺良恶性结节的诊断效能。n 结果:甲状腺良性结节组Tn 2*、D、f值高于恶性组，差异有统计学意义（n P<0.05），Dn *值组间差异无统计学意义（n Z=1.258，n P=0.215）。Tn 2*、D、f值鉴别良恶性结节曲线下面积（AUC）分别为0.703、0.892、0.743，诊断灵敏度分别为80.2%、84.1%、80.0%，特异度分别为65.6%、81.3%、75.0%。Tn 2*联合D、Tn 2*联合Dn *、Tn 2*联合f的AUC分别为0.925、0.709、0.743，诊断灵敏度分别为96.2%、80.4%、80.0%，特异度分别为81.2%、65.6%、75.0%。n 结论:ZOOMit IVIM参数D、f和Tn 2*值在区分甲状腺良恶性结节方面具有很好的价值，其中Tn 2*联合D的诊断效能较高。n “,”Objective:To evaluate the efficacy of Tn 2* mapping and zoomed imaging with parallel transmission technique (ZOOMit) intravoxel incoherent motion (IVIM) DWI in differentiating benign and malignant thyroid nodules.n Methods:The data of 67 patients with thyroid nodules confirmed by surgery and pathology in Ma′anshan People′s Hospital from July 2019 to March 2020 were retrospectively collected. There were a total of 80 nodules, of which 39 were benign nodules and 41 were malignant nodules. All patients underwent MRI including Tn 2WI, Tn 2* mapping, and ZOOMit IVIM sequence scans before surgery to measure the Tn 2*, D, Dn *, and f values of the nodules. The comparison of the parameters between benign and malignant nodules was evaluated by independent sample n t test or Wilcoxon rank sum test. A combined parameter model was established through two-class logistic regression and the predicted probability value was calculated. The ROC curve was used to evaluate the diagnostic efficacy of each parameter and the combined parameter model in differentiating between benign and malignant thyroid nodules.n Results:The Tn 2*, D, f values of benign thyroid nodule group were higher than that in malignant group, and the difference was statistically significant (n P<0.05). The difference of Dn * value between two groups was not statistically significant (n Z=1.258, n P= 0.215). The area under ROC curve (AUC) of Tn 2*, D, and f values for identifying benign and malignant nodules was 0.703, 0.892, and 0.743, the diagnostic sensitivity was 80.2%, 84.1%, and 80.0%, and the specificity was 65.6%, 81.3%, and 75.0%, respectively. The AUC of Tn 2*combined with D, Tn 2*with Dn * and Tn 2* with f was 0.925, 0.709, and 0.743, the diagnostic sensitivity was 96.2%, 80.4%, and 80.0%, and the specificity was 81.2%, 65.6%, and 75.0%, respectively.n Conclusion:D, f and Tn 2* derived from ZOOMit IVIM have good value in differentiating between benign and malignant thyroid nodules, and the combination of Tn 2* and D parameters has the higher diagnostic efficacy.n