Zhang Xing, Tian Jie, Xiang Dehui, Li Xiuli, Deng Kexin
Intelligent Medical Research Center, Institute of Automation, Chinese Academy of Sciences, Beijing 100190, China.
Annu Int Conf IEEE Eng Med Biol Soc. 2011;2011:6005-8. doi: 10.1109/IEMBS.2011.6091484.
In this paper, we present an interactive method for liver tumor segmentation from computed tomography (CT) scans. After some pre-processing operations, including liver parenchyma segmentation and liver contrast enhancement, the CT volume is partitioned into a large number of catchment basins under watershed transform. Then a support vector machines (SVM) classifier is trained on the user-selected seed points to extract tumors from liver parenchyma, while the corresponding feature vector for training and prediction is computed based upon each small region produced by watershed transform. Finally, some morphological operations are performed on the whole segmented binary volume to refine the rough segmentation result of SVM classification. The proposed method is tested and evaluated on MICCAI 2008 liver tumor segmentation challenge datasets. The experiment results demonstrate the accuracy and efficiency of the proposed method so that indicate availability in clinical routines.
在本文中，我们提出了一种从计算机断层扫描（CT）图像中进行肝脏肿瘤分割的交互式方法。经过一些预处理操作，包括肝实质分割和肝脏对比度增强后，在分水岭变换下将CT体数据划分为大量的汇水盆地。然后在用户选择的种子点上训练支持向量机（SVM）分类器，以便从肝实质中提取肿瘤，同时基于分水岭变换产生的每个小区域计算用于训练和预测的相应特征向量。最后，对整个分割后的二值体数据执行一些形态学操作，以细化SVM分类的粗略分割结果。所提出的方法在MICCAI 2008肝脏肿瘤分割挑战赛数据集上进行了测试和评估。实验结果证明了该方法的准确性和效率，表明其在临床常规中的可用性。
