3.0T DCE-MRI及DW-MRI在乳腺良恶性病变诊断中的应用研究

发布时间：2018-05-15 19:19

本文选题：乳腺病变 + 磁共振成像　；参考：《河北医科大学》2014年硕士论文

【摘要】：第一部分3.0T MRI乳腺动态增强曲线半定量分型方法的初步研究目的：探讨3.0T磁共振（MRI）动态增强曲线的半定量分型方法鉴别乳腺良、恶性病变的诊断价值。方法：对72例(81个病灶)女性乳腺疾病患者(平均年龄为48.05±12.59岁)行动态增强扫描。应用半定量方法，采用多个临界值（每秒平均流出斜率的0.03%、每秒平均流出斜率的0.06%、增强初期后信号变化5%及增强初期后信号变化10%）对动态增强曲线进行分型，，并参考Fischer评分标准，分析比较不同的临界值确定的曲线类型联合Fischer评分标准中其它评价指标后用于鉴别诊断良、恶性病变的敏感性、特异性及准确性。结果：每秒平均流出斜率的0.03%、每秒平均流出斜率的0.06%、增强初期后信号变化5%、增强初期后信号变化10%这四个临界值定量评价曲线类型的受试者工作特征曲线(ROC)的曲线下面积（AUC）分别为0.768、0.736、0.764、0.728；由每秒平均流出斜率的0.03%、每秒平均流出斜率的0.06%、增强初期后信号变化5%、增强初期后信号变化10%四个临界值确定的曲线类型结合Fischer评分表中其它指标鉴别良恶性病变的敏感性分别为93.18%、90.91%、88.64%及86.36%；特异性分别为67.57%、59.46%、67.56%及59.46%；准确性分别为81.48%、76.54%、79.01%及74.07%。结论：应用每秒平均流出斜率的0.03%这一临界值划分的动态增强曲线类型最有利于3.0T DCE-MRI对乳腺良恶性病变的鉴别诊断。第二部分3.0T DW-MRI对乳腺局部病变定量的客观性研究目的：通过设计不同的表观扩散系数(ADC值)测量的影响因素，验证3.0T磁共振扩散加权成像（DWI）对乳腺局灶性病变诊断的可重复性及客观性。方法：对53例(61个病灶)女性乳腺疾病患者(平均年龄为47.61±13.09岁)进行DWI扫描。两名观察者分别对不同b值(500，700，1000s/mm2)的DWI图像进行两个感兴趣区（ROI，ROI/2）ADC值的测量；半年后再由观察者A对上述病变的ADC值进行重复测量。结果：1.观察者间及观察者内测量的ADC值差异均无统计学意义(P值分别为0.957，0.547)；两名观察者间所测量ADC值差值平均值为0.00×10-3mm2/s(一致性界限：-0.32×10-3mm2/s-0.32×10-3mm2/s)；同一观察者前、后两次测量的ADC值差值为0.08×10-3mm2/s(一致性界限：-1.26×10-3mm2/s-1.41×10-3mm2/s)。2.在b值相同时，同一观察者对不同ROI大小测量ADC值的差异均无统计学意义（P0.05）。3.不同b值组间比较，ROI组b=500s/mm2组与b=700s/mm2组间、ROI/2组b=500s/mm2组与b=1000s/mm2组间ADC值的差异均具有统计学意义（P0.05），且前者均大于后者。结论：3.0T MRI扩散加权成像ADC值在乳腺病变定性诊断中具有较好的客观性及可重复性，应用同一b值进行疾病诊断更具有临床价值。第三部分3.0T DW-MRI不同b值对乳腺良恶性病变诊断阈值的研究目的：探讨3.0T磁共振(MRI)不同扩散敏感因子（b值）下弥散加权成像(DWI)对良、恶性病变的诊断水平。方法：对53例(61个病灶)女性乳腺疾病患者(平均年龄为47.61±13.09岁)进行DWI扫描，分别在不同b值(500，700，1000s/mm2)下测量病变的ADC值。根据受试者工作特性曲线（ROC）确定不同b值下DWI鉴别诊断良恶性病变的敏感性、特异性以及鉴别良、恶性病变ADC值的最佳阈值。结果：三组b值组中，良性病变与恶性病变ADC值的差异均具有统计学意义(P0.05)，且良性组ADC值均大于恶性组。无论是良性组还是恶性组，三个b值组间的ADC值比较差异并不明显(P0.05)，但经两两比较后发现b=500s/mm2组与b=1000s/mm2组间ADC值存在差异，且差异具有统计学差异(P0.05)。b=500s/mm2时，以1.8×103mm2/s为阈值，鉴别诊断良、恶性病变的敏感度、特异度分别为68%、73%；而b=700s/mm2及b=1000s/mm2时，以ADC阈值分别为1.445×103mm2/s、1.44×103mm2/s，相应鉴别良恶性病变的敏感度、特异度分别为80%、66.7%及72%、80%。结论：DWI对乳腺良恶性病变的鉴别诊断具有重要价值，当b=700s/mm2时有利于病变的检出。
[Abstract]:The first part is a preliminary study on semi quantitative typing of 3.0T MRI dynamic enhancement curve.
Objective: To explore the diagnostic value of semi quantitative typing of 3.0T magnetic resonance (MRI) dynamic enhancement curve in differentiating benign and malignant breast lesions.
Methods: dynamic enhanced scan was performed on 72 cases (81 lesions) of female breast disease (average age of 48.05 + 12.59 years). Semi quantitative method was used, using multiple critical values (0.03% of the average outflow rate per second, 0.06% of the average outflow slope per second, enhanced initial signal change 5%, and enhanced initial signal change 10%). The lines were typed, and the Fischer scoring standard was used to analyze the sensitivity, specificity and accuracy of the diagnosis of benign and malignant lesions after the analysis of the different critical values of the curve types combined with the other evaluation indexes in the Fischer scoring standard.
Results: 0.03% of the average outflow slope per second, 0.06% of the average outflow slope per second, the enhancement of the initial signal change 5%, and the enhancement of the initial signal change 10%, the area (AUC) under the curve of the subjects' working characteristic curve (ROC) of the four critical value curve types is 0.768,0.736,0.764,0.728, respectively, and the average outflow slope per second. 0.03%, 0.06% of the average outflow rate per second, the enhanced initial signal change 5%, the enhanced initial signal change 10% four critical values of the curve type combined with the other indicators in the Fischer score table to identify the sensitivity of benign and malignant lesions were 93.18%, 90.91%, 88.64% and 86.36%, and the specificity were 67.57%, 59.46%, 67.56% and 59.46%, respectively. The accuracy was 81.48%, 76.54%, 79.01% and 74.07%., respectively.
Conclusion: the type of dynamic enhancement curve, which is divided by the 0.03% critical value of the average outflow rate per second, is most beneficial to the differential diagnosis of 3.0T DCE-MRI for benign and malignant breast lesions.
The second part is the objectivity of quantitative analysis of breast lesions by 3.0T DW-MRI.
Objective: to verify the repeatability of 3.0T magnetic resonance diffusion-weighted imaging (DWI) in the diagnosis of focal lesions of the breast by designing different factors of apparent diffusion coefficient (ADC) measurement.
And objectivity.
Methods: 53 cases (61 lesions) of women with breast disease (average age of 47.61 + 13.09 years) were scanned by DWI. Two observers measured the ADC values of the DWI images of different b values (5007001000s/mm2) in two regions of interest (ROI, ROI/2), and then repeated the measurements of the ADC values of the above lesions by the viewer A.
Results: there was no significant difference in the ADC values measured between the 1. observers and the observers (P value was 0.957,0.547); the average value of the ADC value difference between two observers was 0 x 10-3mm2/s (-0.32 x 10-3mm2/s-0.32 x 10-3mm2/s), and the difference value of the ADC value of the latter two measurements was 0.08 * 10-3mm2/s before the same observer. Consistency limits: -1.26 x 10-3mm2/s-1.41 x 10-3mm2/s).2., when the b value is the same, there is no statistically significant difference between the same observer on the measurement of ADC values of different ROI sizes (P0.05).3. (P0.05).3. different b values. P0.05), and the former is larger than the latter.
Conclusion: the ADC value of 3.0T MRI diffusion weighted imaging has good objectivity and repeatability in the qualitative diagnosis of breast lesions. It is of more clinical value to use the same b value to diagnose the disease.
The third part is the study of the diagnostic threshold of 3.0T DW-MRI with different b values for benign and malignant breast lesions.
Objective: To evaluate the diagnostic accuracy of diffusion-weighted imaging (DWI) of 3.0T diffusion weighted imaging (MRI) with different diffusion sensitive factors (b) in benign and malignant lesions.
Methods: 53 cases (61 lesions) of women with breast disease (average age of 47.61 + 13.09 years) were scanned by DWI, and the ADC values were measured at different b values (5007001000s/mm2). The sensitivity, specificity, and differential diagnosis of benign and malignant diseases in the differential diagnosis of benign and malignant lesions under different b values were determined according to the working characteristic curve of the subjects (ROC). The best threshold for changing the ADC value.
Results: in the three groups of B values, the difference of ADC values between benign lesions and malignant lesions were statistically significant (P0.05), and the ADC values in the benign group were all larger than those of the malignant group. The ADC values between the benign groups and the malignant groups were not significant (P0.05), but the ADC values between the b=500s/mm2 and the b=1000s/mm2 groups were found after 22 comparison. When the difference was statistically different (P0.05).B=500s/mm2, 1.8 x 103mm2/s was used as threshold value to identify the sensitivity of benign and malignant lesions, the specificity was 68%, 73% respectively. While b=700s/mm2 and b=1000s/mm2, the ADC threshold was 1.445 * 103mm2/s and 1.44 x 103mm2/s respectively, and the sensitivity of the benign and malignant lesions was identified, and the specificity was 8, respectively. 0%, 66.7% and 72%, 80%.
Conclusion: DWI is of great value in the differential diagnosis of benign and malignant breast lesions. When b=700s/mm2 is helpful for the detection of lesions.

【学位授予单位】：河北医科大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：R445.2;R737.9

【引证文献】