Fully convolutional neural network with post-processing methods for automatic liver segmentation from CT
Yao Zhang, Zhiqiang He, Cheng Zhong, Yang Zhang, Zhongchao Shi · 2017
Automatic liver segmentation from abdominal Computed Tomography (CT) is an important step for hepatic disease diagnosis. It is a challenging task owing to the similarity between liver and its adjacent organs and the low contrast of liver texture (e.g. tumors and blood veins). In this paper, we propose a cascaded structure to automatically segment liver in CT scans. First, we train a fully convolutional neural network (FCN) for coarse liver segmentation; second, we make a comparative study for the performance of different classical segmentation models as the post-processing step to refine the liver segmentation, such as graph cut based method, level set based method and conditional random field (CRF). The main contributions are: (1) the enhancement of FCN for better liver segmentation; (2) the first comparative study on the performance of different classical segmentation models as the post-processing step. Our proposed model is validated on the commonly used database 3DIRCADb, and the experimental results demonstrate that our model excels other models.