基于肺部动态增强MR影像的药物动力学定量分析

针对临床肺部对比剂动态增强磁共振成像(DCE-MRI)数据，提出一种新的药物动力学参数定量分析FCM-BRR方法.该方法基于参考区域模型(referenceregion，RR)，结合模糊C均值聚类(fuzzyC-mean，FCM)和B-样条(B-Spline)曲线拟合技术，解决了传统模型针对具有复杂血供系统的局限性，并且降低了噪声对结果的影响.利用FCM-BRR方法分析了5位病理证实的肺癌患者的临床DCE-MRI数据，检验了该方法的有效性和可行性.FCM-BRR方法给出的Ktrans参数图像反映了肿瘤组织和正常组织的异质性，并通过统计分析知正常组织和肿瘤组织之间具有显著差异(P<001).FCM-BRR方法能够直接利用常规临床数据定量分析病灶区域的药物动力学参数，在治疗评估肺部恶性肿瘤上具有参考意义.

Quantitative Analysis of Pharmacokinetic Parameters Using Clinical Lung Dynamic Contrast Enhanced MR Image

Based on the dynamic contrast enhanced MRI (DCE MRI) data routinely acquired from a clinical lung， a new method called FCM B RR was proposed for quantitative analysis. This method was based on the reference region model (RR)， and combined with fuzzy C mean (FCM) and B Spline curve fitting techniques. The proposed FCM B RR method not only solved the limitation of complex blood supply system， but also reduced the effect of noise on the quantitative analysis result. 5 patients with pathologically proven lung cancer were examined using the proposed FCM B RR method to demonstrate the feasibility and effectiveness. A significant difference (P<001) in Ktransbetween tumor tissue and normal tissue was found. It could be concluded that the proposed FCM B RR method could be used for quantitative analysis of typical clinical lung DCE MRI data， and the Ktrans information has reference value for treatment and assesses the lung cancer.