超声造影参数图像处理在卵巢肿块诊断中的应用

目的探讨超声造影参数图像处理在卵巢肿块诊断中的应用价值。方法2008年1月至2010年12月在武汉大学人民医院行常规超声检查发现的50例卵巢肿块病例，其中良性肿块28例，恶性肿块22例，对其进行常规超声和超声造影检查，应用超声造影参数图像处理软件实时无创动态分析卵巢肿块组织血管的生成和血流灌注模式，比较其时间一强度曲线（TIC）和动态血管模式曲线（DVP）的形态特征，并对TIC各项定量指标进行分析。结果卵巢肿块的超声造影图像中，良性肿块有24例（86％）显示内部以蓝色为主的低增强，恶性肿块有15例（68％）显示内部以红色为主的高增强，差异有统计学意义（P〈0．01）；良性肿块中，23例（82％）DVP曲线在造影全过程全部时段保持在基线之下，恶性肿块中有15例（68％）DVP曲线在造影全过程全部时段保持在基线之。卜；造影剂达峰强度恶性肿块显著高于良性肿块（P=0．000），达峰时间及平均渡越时间恶性肿块短于良性肿块（28±11）S比（65±19）S，（74±29）S比（154±52）S]，差异均有统计学意义（均P〈0．05）。始增时间恶性肿块短于良性肿块（23±10）S比（44±11）S]，差异有统计学意义（P〉0．05）。结论卵巢肿块的超声造影TIC、DVP曲线在良恶性之间存在明显差异，而结合了超声造影参数图像处理软件可更加直观地反映良恶性肿块的定量信息。

Application of parametric image processing of contrast-enhanced ultrasound imaging in diagnosis of ovarian masses

Objective To investigate application value of parametric image processing in contrastenhanced ultrasound imaging in diagnosis of ovarian masses. MethodsFifty cases with ovarian masses underwent routine ultrasound and contrast-enhanced ultrasound imaging using a new dedicated parametric image processing software SonoLiver to analyze patterns of vascular formation and blood stream perfusion in the ovarian mass tissues on a digital video recorder in real time, compare their morphological characteristics of time-intensity curve (TIC) and dynamic vascular patterns (DVP) curve, and analyze quantitatively all indicators generated by SonoLiver. ResultsIn ultrasound imaging of the 50 cases, there were 24 cases (86％) displaying mainly blue lowly-enhanced imaging in those with benign masses and 15 cases (68％)displaying mainly red highly-enhanced imaging in those with malignant masses, with statistical significance (P ＜0. 01 ). There was significantly different characteristics of TIC and DVP between patients with benign and malignant masses. In 23 cases with benign masses, their DVP were significantly higher above the baseline than in those with malignant ones ( P = 0. 000), and in 15 cases with malignant lesions, their DVP were much shorter below the baseline than in those with benign ones, with statistical significance (P ＜0.05). The intensity of contrast medium, the time to reach its peak intensity and average transit time were all significantly higher in those with malignant masses than in those with benign ones (all P ＜ 0. 05 ). But, no statistical difference in the time of initial increasing between the two groups was found (P ＞ 0. 05). Conclusions There is significant difference in TIC and DVP of ultrasound imaging between benign and malignant masses, which if combined with contrast enhanced ultrasound parametric image processing can provide a more visualized quantitative information of benign and malignant ovarian masses with SonoLiver software.