首页 | 本学科首页   官方微博 | 高级检索  
     

磁共振动态对比增强直方图分析在前列腺癌鉴别诊断中的价值
引用本文:余加懿,郭大静,罗银灯,何晓静,凌茜,吴伟,陈维娟,白珊玮. 磁共振动态对比增强直方图分析在前列腺癌鉴别诊断中的价值[J]. 第三军医大学学报, 2017, 39(22). DOI: 10.16016/j.1000-5404.201708017
作者姓名:余加懿  郭大静  罗银灯  何晓静  凌茜  吴伟  陈维娟  白珊玮
作者单位:重庆医科大学附属第二医院放射科,重庆,400010
基金项目:国家自然科学基金青年科学基金项目,重庆市卫计委医学科研计划项目(2016MSXM024
摘    要:目的 探讨磁共振动态对比增强(dynamic contrast enhanced magnetic resonance imaging, DCE-MRI)定量参数直方图分析对前列腺癌检出的应用价值.方法 收集本院2016年10月至2017年6月行盆腔常规MRI、DWI和DCE-MRI扫描的患者72例,经病理证实的前列腺癌(prostate cancer,PCa)38例,前列腺增生(prostatic hyperplasia,BPH)18例,结合影像表现、实验室检查及随访观察诊断前列腺炎16例.利用第三方后处理软件(Omni Kinetic,OK),在病灶最大中心层面勾画感兴趣区,获得感兴趣区内DCE-MRI定量参数(Ktrans、Kep和Ve)的平均值及各百分位数值(5%、10%、25%、50%、75%、90%和95%),分别比较外周带PCa与前列腺炎、中央腺体PCa与BPH各参数值的差异,并分析诊断效能.结果 外周带PCa组Ktrans平均值及各百分位数值均大于前列腺炎(平均值及50%、75%、90%和95%百分位数值P<0.05).PCa组Kep平均值及各百分位数值均大于前列腺炎(P<0.05).外周带PCa组Ve平均值及各百分位数值小于或等于前列腺炎(5%、10%、25%和50%百分位数值P<0.05).中央腺体PCa与BPH渗透参数Ktrans、Kep、Ve的平均值及各百分位数值差异均无统计学意义(P>0.05).外周带PCa组Ktrans和Kep高百分位数AUC大于平均值,低百分位数AUC最小,90% Kep诊断效能最大(AUC为0.932,敏感度78.3%,特异度92.9%);外周带PCa组Ve百分位数不具有诊断效能.结论 基于DCE-MRI的直方图分析在外周带PCa与前列腺炎鉴别诊断中具有临床价值,Ktrans、Kep高百分位数诊断效能优于平均值和低百分位数.

关 键 词:前列腺肿瘤  磁共振动态对比增强  直方图分析

Value of histogram analysis of quantitative parameters in dynamic contrast-enhanced magnetic resonance imaging in diagnosis of prostate cancer
YU Jiayi,GUO Dajing,LUO Yindeng,HE Xiaojing,LING Qian,WU Wei,CHEN Weijuan,BAI Shanwei. Value of histogram analysis of quantitative parameters in dynamic contrast-enhanced magnetic resonance imaging in diagnosis of prostate cancer[J]. Acta Academiae Medicinae Militaris Tertiae, 2017, 39(22). DOI: 10.16016/j.1000-5404.201708017
Authors:YU Jiayi  GUO Dajing  LUO Yindeng  HE Xiaojing  LING Qian  WU Wei  CHEN Weijuan  BAI Shanwei
Abstract:Objective To investigate the value of histogram analysis of the quantitative parameters obtained by dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) in the diagnosis of prostate cancer (PCa).Methods Thirty-eight patients with pathologically confirmed PCa and 18 patients with benign prostatic hyperplasia (BPH) were enrolled in this study along with 16 patients with prostatitis diagnosed by radiographic presentations,laboratory findings and follow-up observation.All the patients underwent conventional pelvic MRI,diffusion-weighted imaging (DWI) and DCE-MRI examinations between October 2016 and June 2017,and histogram analysis was performed using a third-party post-processing software (Omni Kinetic,OK).Regions of interest (ROIs) were drawn on the largest central level of the lesions,and the software automatically analyzed the mean,5%,10%,25%,50%,75%,90% and 95% percentiles of DCE-MRI quantitative parameters (Ktrans,Kep,and Ve) in the ROIs.Each of the parameters was compared between PCa and prostatitis in the peripheral zone and between PCa and BPH in the central gland zone to analyze their respective diagnostic efficacy.Results In the peripheral zone,the mean and percentiles (50%,75%,90% and 95%) of Ktrans and Kep in PCa were significantly higher than those in prostatitis (P <0.05);the 5%,10%,25% and 50% percentiles of Ve in PCa were significantly smaller than those in prostatitis (P < 0.05).In the central gland zone,the mean and percentiles of Ktrans,Kep and Ve all showed no significant differences between PCa and BPH.In PCa in the peripheral zone,the area under the curves (AUC) of high percentiles of Ktrans and Kep were greater those of the means,with the AUCs of the low percentiles being the smallest.The diagnostic efficacy of 90% Kep was the highest for PCa in the peripheral zone with an AUC of 0.932,a sensitivity of 78.3%,and a specificity of 92.9%.The percentiles of Ve did not show any diagnostic efficacy for PCa in the peripheral zone.Conclusion Histogram analysis of the quantitative parameters of DCE-MRI has a clinical value in the differential diagnosis between peripheral PCa and prostatitis.The high percentiles of Ktrans and Kep have better diagnostic efficiencies than the mean and low percentiles of Ktrans and Kep.
Keywords:prostatic neoplasms  dynamic contrast-enhanced magnetic resonance imaging  histogram analysis
本文献已被 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号