Abstract
The liver is an essential metabolic organ of the human body, and malignant liver tumors seriously affect and threaten human life. The segmentation algorithm for liver and liver tumors is one of the essential branches of computer-aided diagnosis. This paper proposed a two-stage liver and tumor segmentation algorithm based on the convolutional neural network (CNN). In the present study, we used two stages to segment the liver and tumors: liver localization and tumor segmentation. In the liver localization stage, the network segments the liver region, adopts the encoding–decoding structure and long-distance feature fusion operation, and utilizes the shallow features’ spatial information to improve liver identification. In the tumor segmentation stage, based on the liver segmentation results of the first two steps, a CNN model was designed to accurately identify the liver tumors by using the 2D image features and 3D spatial features of the CT image slices. At the same time, we use the attention mechanism to improve the segmentation performance of small liver tumors. The proposed algorithm was tested on the public data set Liver Tumor Segmentation Challenge (LiTS). The Dice coefficient of liver segmentation was 0.967, and the Dice coefficient of tumor segmentation was 0.725. The proposed algorithm can accurately segment the liver and liver tumors in CT images. Compared with other state-of-the-art algorithms, the segmentation results of the proposed algorithm rank the highest in the Dice coefficient.
Highlights
With the development of computer technology, computer-aided technology has been a popular method to analyze medical images, which can assist clinicians in detecting and segmenting tumor lesion regions from normal tissues
We presented several solutions: (1) we designed a two-stage densely connected UNet (DCUNet) for liver and liver tumor segmentation, which consists of two stages, and we focused on both 2D and 3D features in the proposed algorithm; and (2) we added an attention mechanism to the neural network architecture to learn the multi-scale features of small tumors in the liver
All CT images used in this experiment are from the Liver Tumor Segmentation Challenge of the 2017 International Conference on Medical Image Computing and ComputerAssisted Intervention (MICCAI)
Summary
With the development of computer technology, computer-aided technology has been a popular method to analyze medical images, which can assist clinicians in detecting and segmenting tumor lesion regions from normal tissues. Aqyyum et al [9] proposed a 3D hybrid model for CT images, which consisted of a three-dimensional residual network, spatial squeeze module, and excitation module This algorithm performed well for the segmentation of liver and large tumor regions, but the detection of small tumor regions was not accurate. The existing algorithms performed well in segmenting liver and liver tumors, there are still some shortcomings: (1) they focus on either 2D features or 3D features of the liver CT images, and ignore the hybrid features from 2D and 3D; and (2) segmentation performance of small liver tumors is poor, which is caused by the small proportion of small liver tumor in the CT image and low gradient between the liver tumor and background To address these shortcomings, we presented several solutions: (1) we designed a two-stage densely connected UNet (DCUNet) for liver and liver tumor segmentation, which consists of two stages, and we focused on both 2D and 3D features in the proposed algorithm; and (2) we added an attention mechanism to the neural network architecture to learn the multi-scale features of small tumors in the liver
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
