Automatic segmentation of liver tumors from multiphase contrast-enhanced CT images based on FCNs.

•The multi-channel fully convolutional networks is designed.•We segment liver tumors from multiphase contrast-enhanced CT images.•We train one network for each phase of CT images and fuse their high-layer features together.•This method can make full use of the characteristics of different enhancement phases of CT images.•The results showed our model provided greater accuracy and robustness than previous methods.