Ultrasonic Liver Tissues Classification for Radiofrequency Tumor Ablation Treatment Evaluation
Ultrasonography is one of the safest methods used as an effective diagnostic tool. The non-radioactive attribute makes it stand out in evaluating cancer treatment when the application of CT and MRI may have deteriorating effects. In this paper, for ultrasonic liver image analysis, approaches based on Multiresolution fractal feature vector is explored to detect residual tumor tissue after the treatment of Radio Frequency Ablation (RFA). This technique is applied on three sets of ultrasonic liver images, normal, cancer, and post-treatment coagulation necrosis in different periods, all histologically proven. In all images, 32*32 pixel rectangular regions of interest were selected by specialized physicians and used in the analysis. Our experiment demonstrates the feasibility of differentiation residual tumor from post-treatment necrosis and normal tissue by using Multiresolution fractal (MF) feature vector, which may become a promising auxiliary tool for clinic evaluation of the RAF treatment effects and guidance providing for prospective therapy.
Ultrasound liver Radiofrequency ablation (RFA) multiresolution fractal feature vector
Shen Sun Xinhua Xiao Fang Li Jingjing Guo Ying Liu Tianshuang Qiu
Department of Electronic Engineering,Dalian University of Technology,116024 Dalian,China School of Information Engineering,Northeastern University,110004 Shenyang,China Ultrasound department,The Second Affiliated Hospital of Dalian Medical University,116027 Dalian,Chin
国际会议
北京
英文
1-4
2009-06-11（万方平台首次上网日期，不代表论文的发表时间）