杨晓彤,刘屹,黄德生,赵宇,李娇,任克.肾脏肿瘤能谱CT曲线拉伸指数模型定量分析[J].中国医学影像技术,2016,32(7):1092~1096
肾脏肿瘤能谱CT曲线拉伸指数模型定量分析
Quantitative analysis of spectral CT curve stretched-exponential model in kidney tumors
投稿时间:2015-12-03  修订日期:2016-05-05
DOI:10.13929/j.1003-3289.2016.07.028
中文关键词:  能谱成像  肾肿瘤  拉伸指数模型
英文关键词:Spectral imaging  Renal tumour  Stretched-exponential analysis model
基金项目:辽宁省科学技术计划项目(2013225049)
作者单位E-mail
杨晓彤 中国医科大学附属第一医院放射科, 辽宁 沈阳 110001  
刘屹 中国医科大学附属第一医院放射科, 辽宁 沈阳 110001 liuyicmu@sina.cn 
黄德生 中国医科大学公共基础学院数学教研室, 辽宁 沈阳 110122  
赵宇 中国医科大学附属第一医院放射科, 辽宁 沈阳 110001  
李娇 中国医科大学附属第一医院放射科, 辽宁 沈阳 110001  
任克 中国医科大学附属第一医院放射科, 辽宁 沈阳 110001  
摘要点击次数: 1669
全文下载次数: 696
中文摘要:
      目的 采用拉伸指数非线性回归模型对能谱曲线进行解析,探讨其对不同病理类型肾肿瘤的鉴别诊断价值。方法 回顾性分析经手术病理证实并行能谱CT检查的59例肾肿瘤患者影像资料,其中,乏脂肪错构瘤8例(A组),乏血供肾癌17例(B组),肾透明细胞癌(ccRCC)Ⅱ级34例(C组)。随机抽取23例患者,以健侧肾皮质(D组)、腰大肌(E组)作为正常对照。测量并计算病灶实性部分及正常对照的标准化碘浓度(NIC),提取能谱曲线数据并利用拉伸指数非线性回归模型计算拉伸指数(α值)及b值。比较平扫及增强后各组间NIC、α、b值的差异,评估病灶α、b值与NIC的相关性。结果 平扫5组间α值的差异均无统计学意义(P均>0.05)。动脉期A组与B组间NIC及α值差异均无统计学意义(P均>0.05),实质期两者α值差异有统计学意义(P<0.05),但NIC差异均无统计学意义(P>0.05)。动脉期C组与D组各参数差异均无统计学意义(P均>0.05),实质期两组间NIC、α值差异有统计学意义(P均<0.05)。动脉和实质期E组与其他各组NIC及α值的差异均有统计学意义(P均<0.05)。平扫、动脉期及实质期,E组与其他各组间b值差异均有统计学意义(P均<0.05);平扫、实质期,C组与D组间b值差异有统计学意义(P均<0.05),其余各组间b值差异均无统计学意义(P均>0.05)。动脉期(r=0.771)及实质期(r=0.695)病变α值与NIC呈正相关(P均<0.001)。结论 采用拉伸指数非线性回归模型解析能谱曲线对肾脏肿瘤性病变的诊断有一定的临床应用价值。
英文摘要:
      Objective To explore the value of spectral CT in differential diagnosis of kidney tumors using the stretched-exponential model. Methods Totally 59 patients with kidney tumors proved by operation and pathology were retrospectively analyzed, all of them underwent spectral CT in plane and double-phasic enhanced scanning, i.e. 8 minimal fat renal angiomyolipoma (group A), 17 hypovascular renal carcinoma (group B) and 34 clear cell renal cell carcinoma (group C), and then 23 patients were randomly selected, whose the normal renal cortex (group D) and psoas major (group E) as the control. The normalized iodine concentrations (NIC) of the lesion, normal renal cortical and psoas major were measured and calculated. The data of spectral curve were extracted and used to calculate the stretched-exponential index (α) and b value with the stretched-exponential nonlinear regression analysis model. The α, b and NIC values were compared among 5 groups. The correlation between NIC and α, b value of the lesion were analyzed. Results The α value had no differences among all groups in plane scanning (all P>0.05). In nephpographic phase (NP) there were significantly different of α value between group A and group B (P<0.05), but in cortex phase (CP) and NP, there were no differences of NIC between the two groups (all P>0.05). In NP there were significantly different of NIC and α value between group C and group D (all P<0.05). In CP and NP, there were significantly different between group E and other groups in NIC and α value (all P<0.05). For b value, there were significantly different between group E and other groups in plane and enhancement scanning (all P<0.05), as well as group C and group D in plane and NP scanning (all P<0.05). But no significant difference between other groups. The α values of lesions had positive correlation with NIC in CP and NP (r=0.771, 0.695, both P<0.001). Conclusion Quantitative spectral CT with the stretched-exponential nonlinear regression analysis model has some application value in differential diagnosis of renal tumor.
查看全文  查看/发表评论  下载PDF阅读器