多b值DWI和DCE-MRI定量分析联合应用对卵巢肿瘤的诊断价值研究

发布时间：2018-03-12 08:18

本文选题：卵巢肿瘤　切入点：DWI　出处：《福建中医药大学》2017年硕士论文　论文类型：学位论文

【摘要】：目的探讨多b值弥散加权成像、DCE-MRI及两者联合应用对诊断卵巢肿瘤的价值及可行性,为卵巢肿瘤的良、恶性鉴别提供无创、可复性及一定的参考依据。方法收集40例经手术病理证实为卵巢肿瘤病例的MRI检查资料,所有患者术前均采用SIEMENS Verio 3.0T超导磁共振扫描仪行常规MRI平扫、多b值DWI及DCE-MRI检查,分析病变的影像学特征作出定性诊断,与术后组织病理结果进行对照得出敏感性(Se)、特异性(Sp)、阳性预测值(PPV)、阴性预测值(NPV)及准确度;其次分别测量卵巢肿瘤实性与囊性部分的ADC值,得出卵巢肿瘤实性部分的TIC曲线及相对应的定量参数Ktrans值、Kep值及Ve值,分别对上述数据进行统计分析。最后,利用ROC曲线分析得出各个参数值对鉴别卵巢肿瘤良、恶性的最优参数值及最佳诊断阈值,及相对应的敏感性、特异性、阳性预测值及阴性预测值。结果(1)多b值DWI联合DCE-MRI检查对卵巢肿瘤的诊断效能较单一检查方式高,敏感性为94.7%,特异性为100%,阳性预测值为100%,阴性预测值为95.5%,准确度97.5%;(2)两组卵巢肿瘤病例的年龄组间差异无统计学意义(P0.05),不同类型的TIC曲线在卵巢肿瘤良、恶性的鉴别上具有显著的统计学意义(P0.01);(3)卵巢良、恶性肿瘤实性部分的各组间ADC值、Ktrans值、Kep值及Ve值差异均具有统计学意义(P0.05);囊性部分的各组间ADC值差异无统计学意义(P0.05);(4)利用ROC曲线分析,参考Youden指数的最大值得出诊断卵巢肿瘤良、恶性的临界值,当b值为1200s/mm2时具有最佳诊断效能,ADC值的AUC为0.937,阈值为1.17×10-3mm2/s,敏感性为90.5%,特异性为89.5%,阳性预测值为89.6%,阴性预测值为90.4%;DCE-MRI定量分析中定量参数Ktrans值具有更高的诊断效能,Ktrans值的AUC为0.857,最佳诊断阈值为0.093min-1,敏感性为85.7%,特异性为72.2%,阳性预测值为75.5%,阴性预测值为83.5%。结论(1)b值为1200s/mm2是鉴别卵巢肿瘤良、恶性的最佳扫描参数,ADC值对卵巢肿瘤的定性具有肯定的诊断价值;(2)Ktrans值是DCE-MRI定量分析中对卵巢肿瘤良、恶性诊断效能最高的定量参数值;(3)多b值DWI联合DCE-MRI扫描提高了卵巢肿瘤术前诊断的准确性。
[Abstract]:Objective to evaluate the value and feasibility of DCE-MRI and DCE-MRI in the diagnosis of ovarian neoplasms, so as to provide noninvasive differential diagnosis between benign and malignant ovarian tumors. Methods 40 cases of ovarian neoplasms proved by operation and pathology were examined by MRI. All patients were examined by SIEMENS Verio 3.0T superconducting magnetic resonance scanner (SIEMENS Verio 3.0T) before operation, including routine MRI scan, multi-b value DWI and DCE-MRI examination. The qualitative diagnosis was made by analyzing the imaging features of the lesions, and the sensitivity, specificity, positive predictive value and negative predictive value of NPV were obtained by comparing with the postoperative histopathological results. Secondly, the ADC values of solid and cystic parts of ovarian tumors were measured, and the TIC curves of solid parts of ovarian tumors and the corresponding quantitative parameters, Ktrans value and ve value, were obtained, and the above data were statistically analyzed. By using ROC curve analysis, the optimal parameter values and diagnostic thresholds for differentiating benign and malignant ovarian tumors were obtained, as well as their sensitivity and specificity. Positive predictive value and negative predictive value. Results: Multi-b DWI combined with DCE-MRI was more effective in the diagnosis of ovarian tumor than that in single mode. The sensitivity was 94.7 and the specificity was 100. The positive predictive value was 100. The negative predictive value was 95.5and the accuracy was 97.50.There was no significant difference between the two groups in the age of ovarian neoplasms (P 0.05). The TIC curves of different types were good in ovarian tumors. The differentiation of malignancy was statistically significant (P 0.01). There were significant differences in ADC value, Ktrans value, Kep value and ve value among groups in solid part of malignant tumor (P 0.05), but there was no significant difference in ADC value among groups in cystic part (P 0.05). Referring to the maximum value of Youden index, the critical value for the diagnosis of benign and malignant ovarian tumors is obtained. When b value is 1200s / mm2, the AUC with the best diagnostic efficiency and ADC value is 0.937, the threshold is 1.17 脳 10 ~ (-3) mm ~ (-2) / s, the sensitivity is 90.5 and the specificity is 89.5, the positive predictive value is 89.6 and the negative predictive value is 90.4 / DCE-MRI quantitative analysis, the quantitative parameter Ktrans value has higher diagnostic efficacy and Ktrans value. AUC was 0.857, the best diagnostic threshold was 0.093 min-1, the sensitivity was 85.7, the specificity was 72.2, the positive predictive value was 75.5 and the negative predictive value was 83.5.Conclusion the best diagnostic threshold is 0.92 min -1, the sensitivity is 85.7 min, the specificity is 72.2%, the positive predictive value is 75.5 and the negative predictive value is 83.5%. The best scanning parameter of malignant tumor and the value of ADC have positive diagnostic value for the qualitative analysis of ovarian tumor. The Ktrans value is good for ovarian tumor in quantitative analysis of DCE-MRI. Multiple b DWI combined with DCE-MRI enhanced the accuracy of preoperative diagnosis of ovarian tumors.
【学位授予单位】：福建中医药大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：R737.31;R445.2

【参考文献】