宝石能谱CT与常规超声对甲状腺结节良恶性诊断价值的对照研究

发布时间：2018-01-06 15:07

本文关键词：宝石能谱CT与常规超声对甲状腺结节良恶性诊断价值的对照研究　出处：《郑州大学》2017年硕士论文　论文类型：学位论文

【摘要】：背景与目的探讨宝石能谱CT对鉴别甲状腺结节良恶性的诊断价值,联合病理结果与常规超声进行对照研究。材料与方法对本院2015年8月至2016年8月行能谱CT扫描并且经过病理证实的甲状腺结节患者55例,男17例,女38例,平均年龄为(54.6±1.0)岁,共计80个结节。根据病理结果将其分为良性组及恶性组两组,并选取35名颈部非甲状腺疾病患者作为正常对照组。所有患者均进行甲状腺及双侧淋巴结超声检查及颈部宝石能谱CT(GE Discovery HD 750 CT扫描仪)增强扫描,并将CT平扫及增强扫描图像及数据传送至ADW4.6工作站进行处理,应用GSI Viewer软件在甲状腺结节病灶实性成分及其周围正常组织放置感应区(Region of interest,ROI),应尽量因避免锁骨产生的伪影区域,若密度不均匀时,应避开囊变、出血及钙化区域。在由后处理生成的碘基图上记录病灶区域实性成分及周围正常甲状腺组织相同感应区内的碘浓度。在增强扫描时,为消除造影剂注入及个体循环因素不同所造成的影响,应计算其在动脉期、静脉期的标准化碘浓度(NIC),公式为:标准化碘浓度=病灶实性部分ROI的碘浓度/同期相颈动脉ROI的碘浓度。由平扫、动脉期及静脉期碘浓度可计算其能谱曲线斜率,本研究选取40及100ke V为参考点,计算公式为:能谱曲线斜率=(HU40ke V-HU100ke V)/60。通过观察70ke V时获得的单能量图像及碘基图,分别记录甲状腺结节的边界、密度(有无囊变)、钙化形态及部分结节内伴有乳头状结构等。通过分析纳入患者的超声声像图分别记录结节的形态、边界、有无囊变、周围声晕等参数。将能谱CT获得的数据通过单因素方差分析及q检验分别比较平扫、动脉期及静脉期碘浓度及能谱曲线斜率的差异、动脉期及静脉期标准化碘浓度之间的差异。超声图像数据组采用多因素Logistic回归分析。根据两组数据分别绘制出受试者工作特征(ROC)曲线,计算出相应的敏感度及特异度,评价效能。结果1.一般资料55例患者纳入研究,男17例、女38例,年龄21-77岁,平均(54.6±1.0)岁,平均BM=18.1±3.2kg/m2。共80个结节,其中良性结节45例,均为结节性甲状腺肿;恶性结节35例,包括有甲状腺乳头状癌31例,滤泡状癌2例,髓样癌1例,未分化癌1例。正常组35例,男15例、女20例,平均(61.2±7.2)岁,平均BMI(19.9±3.0)kg/m2。2.1甲状腺结节的能谱CT表现2.1.1碘浓度在平扫、动脉期及静脉期,正常组、良性组及恶性组的碘浓度两两比较均具有统计学意义(表1,P0.05)。2.1.2在动脉期,正常组、良性组及恶性组的标准化碘浓度分别为0.84±0.22、0.43±0.14、0.23±0.15,两两比较均有统计学意义(P0.05);静脉期,正常组、良性组及恶性组的标准化碘浓度分别为1.13±0.23、0.64±0.15、0.47±0.18,两两比较均有统计学意义(P0.05)。2.1.3能谱曲线能谱曲线为下降型,斜率是负值。随着keV的值升高,结节区域对应的CT值呈现一个递减的趋势,且CT值递减的幅度逐渐减小。平扫期、动脉期及静脉期能谱曲线斜率差异两两比较具有统计学意义(表2,P0.05)。2.1.4与病理结果对照,仅通过形态学影像诊断甲状腺恶性结节的敏感度和特异度分别为66.42%和79.49;在静脉期标准化碘浓度区小于0.56和能谱曲线斜率绝对值小于1.48为阈值诊断恶性结节时,与病理结果具有较好的一致性,通过在静脉期标准化碘浓度区小于0.56及静脉期斜率绝对值小于1.48为阈值分别诊断甲状腺恶性结节的敏感度和特异度分别为79.6%、78.8%和83.3%、81.10%。联合形态学、碘浓度、标准化碘浓度及能谱曲线多参数分析,诊断甲状腺结节良恶性的敏感度为91.4%,特异度为93.3%。2.2甲状腺结节的超声表现(表3)良性组与恶性组的结节形态、微钙化、声晕及血流形态特征具有统计学意义(P0.05)。回归方程如下:logit(P)=-3.076+形态×1.965+微钙化×2.996+声晕×2.679-Ⅰ级血流×2.174-Ⅱ级血流×2.645+Ⅲ级血流×0.197(表4),将回归方程计算P≥0.50判断为恶性,P0.50判断为良性作为判断标准时,敏感度为77.1.%,特异度为68.9%,曲线下面积为0.757,95%置信区间为(0.643,0.870)。2.3能谱CT与常规超声比较(表5)静脉期联合形态学、标准化碘浓度及能谱曲线多参数分析诊断甲状腺结节的敏感度及特异度均优于常规超声。结论能谱CT及超声检查在鉴别甲状腺结节良恶性方面的应用均表现出了各自的价值。甲状腺良恶性结节的碘浓度及能谱曲线斜率不同,本次研究显示在甲状腺结节直径"g1cm时,静脉期联合形态学、标准化碘浓度及能谱曲线斜率有助于提高甲状腺结节良恶性鉴别的准确率,其敏感度及特异度要优于超声。联合应用能谱CT技术多参数成像可为甲状腺结节的定性诊断提供可靠的影像学数据。
[Abstract]:Background and objective: To investigate the value of spectral CT differential diagnosis of benign and malignant thyroid nodules, were studied with conventional ultrasound combined with pathological results. Materials and methods of spectral CT scan after 55 thyroid nodules were pathologically confirmed cases in our hospital from August 2015 to August 2016, 17 cases of male, female 38 cases, average age (54.6 + 1) years old, a total of 80 nodules. According to the pathological results will be divided into two groups of benign and malignant group, and selected 35 non cervical thyroid disease patients as control group. All patients underwent thyroid ultrasonography and bilateral lymph nodes and neck of gemstone CT (GE Discovery HD 750 CT scanner) enhanced scan and CT scan and enhanced scan image data is transmitted to the ADW4.6 workstation and processed using GSI Viewer software in the sense of thyroid nodules placed solid components and their surrounding tissues Should (Region of interest, ROI District), should as far as possible due to artifacts generated if the area from the clavicle, uneven density, should avoid cystic degeneration, hemorrhage and calcification area. In the generation process by recording kitu iodine solid lesions component and normal thyroid tissue around the same iodine concentration in the induction zone in enhanced scan, in order to eliminate the influence of contrast agent injection and individual circulation caused by different factors, should be calculated in the arterial phase, venous phase of the normalized iodine concentration (NIC), the formula is: iodine concentrations at the same phase of carotid ROI iodine concentration / standard iodine concentration = the solid portion of the tumor ROI. By scan, the iodine concentration of arterial and venous phase can be calculated and the energy spectrum curve slope, this study selected 40 and 100ke V as a reference point, the formula for calculating the energy spectrum curve slope (HU40ke = V-HU100ke V) single energy /60. by observing the 70ke obtained during V and ghitu like iodine respectively. Record of thyroid nodule boundary, density (with or without cystic degeneration), calcification and part of nodules with papillary structure. Through the analysis into the nodule morphology, were recorded in patients with sonographic boundary has no cystic acoustic halo around other parameters. The spectral CT data acquired by the single factor analysis of variance and Q test were compared with plain scan, arterial phase and venous phase difference of iodine concentration and energy spectrum curve slope, the difference between the arterial and venous phase normalized iodine concentration. Ultrasound image data using multivariate Logistic regression analysis. According to the data of the two groups were drawn by the receiver operating characteristic (ROC) the curve, calculate the corresponding sensitivity and specificity, efficiency evaluation. Results of the 1. general data of 55 patients were enrolled in this study, male 17 cases, female 38 cases, age 21-77 years, average (54.6 + 1) years old, the average BM=18.1 + 3.2kg/m2. 80 nodules, including 45 benign nodules Cases were nodular goiter; 35 cases of malignant nodules, including 31 cases of thyroid papillary carcinoma, 2 cases of follicular carcinoma, 1 cases of medullary carcinoma, undifferentiated carcinoma in 1 cases. 35 cases of normal group, male 15 cases, female 20 cases, average (61.2 + 7.2) years old, an average of BMI (19.9 + 3) CT 2.1.1 spectroscopy showed the iodine concentration in the plain kg/m2.2.1 of thyroid nodules, arterial and venous phase, the normal group, the iodine concentration of 22 benign group and malignant group were statistically significant (Table 1, P0.05.2.1.2) in the arterial phase, normal group, benign group and normalized iodine concentration malignant group were 0.84 + 0.22,0.43 + 0.14,0.23 + 0.15, 22 were statistically significant (P0.05); venous phase, the normal group, the normalized iodine concentration in benign group and malignant group were 1.13 + 0.23,0.64 + 0.15,0.47 + 0.18, 22 were statistically significant (P0.05.2.1.3) can compose energy spectrum curve drop type, with the slope is negative. The keV value increased, CT value corresponding to the lesion area showed a decreasing trend, and the CT value decline rate gradually decreased. The unenhanced phase, arterial phase and venous phase compared to the slope of spectral curve between the 22 was statistically significant (Table 2, P0.05) the control results of.2.1.4 and pathology, only through morphological imaging diagnosis malignant thyroid nodules the sensitivity and specificity were 66.42% and 79.49; in the vein of the normalized iodine concentration area is less than 0.56 and less than 1.48 for the diagnosis of malignant nodules threshold spectrum curve of the absolute value of the slope, has good consistency with pathological results, the normalized iodine concentration in venous phase and venous phase area is less than 0.56 the absolute value of the slope is less than the threshold value of 1.48 respectively in diagnosing malignant thyroid nodules the sensitivity and specificity were 79.6%, 78.8% and 83.3%, 81.10%. combined with morphology, iodine concentration, normalized iodine concentration and energy spectrum analysis of multi parameter curve The diagnosis of benign and malignant thyroid nodules, the sensitivity was 91.4%, specificity of 93.3%.2.2 ultrasonography of thyroid nodules (Table 3) benign group and malignant group of the nodular morphology, micro calcification, significant acoustic halo and flow patterns (P0.05). The regression equation is as follows: logit (P) =-3.076+ * 1.965+ morphology of micro calcification * 2.996+ * 2.679- halo grade II * * 2.174- blood flow 2.645+ III blood flow by 0.197 (Table 4), the regression equation P = 0.50 for the judgment of malignant, benign P0.50 judgment as the criterion, the sensitivity is 77.1.%, the specificity was 68.9%, the area under the curve was 0.757,95% confidence interval (0.643,0.870).2.3 spectroscopy CT and conventional ultrasound (Table 5) combined with venous phase morphology, the normalized iodine concentration and spectrum curve of multi parameter analysis in the diagnosis of thyroid nodules in both sensitivity and specificity than conventional ultrasound. Conclusion CT and ultrasound in energy spectrum The application of differential diagnosis of benign and malignant thyroid nodules showed their value. The iodine concentration of benign and malignant thyroid nodules and energy spectrum curve slope is different, the study showed that the diameter of thyroid nodules "g1cm, combined with venous phase morphology, the normalized iodine concentration and to compose line slope is helpful to improve the accuracy of differential diagnosis of benign malignant thyroid nodules, the sensitivity and specificity is superior to ultrasound. Combined application of spectral CT technology for multi parameter imaging can provide reliable images for diagnosis of thyroid nodule data.

【学位授予单位】：郑州大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：R445.1;R581;R816.6

【相似文献】