Three-Dimensional Liver Image Segmentation Using Generative Adversarial Networks Based on Feature Restoration

Medical imaging provides a powerful tool for medical diagnosis. In the process of computer-aided diagnosis and treatment of liver cancer based on medical imaging, accurate segmentation of liver region from abdominal CT images is an important step. However, due to defects of liver tissue and limitations of CT imaging procession, the gray level of liver region in CT image is heterogeneous, and the boundary between the liver and those of adjacent tissues and organs is blurred, which makes the liver segmentation an extremely difficult task. In this study, aiming at solving the problem of low segmentation accuracy of the original 3D U-Net network, an improved network based on the three-dimensional (3D) U-Net, is proposed. Moreover, in order to solve the problem of insufficient training data caused by the difficulty of acquiring labeled 3D data, an improved 3D U-Net network is embedded into the framework of generative adversarial networks (GAN), which establishes a semi-supervised 3D liver segmentation optimization algorithm. Finally, considering the problem of poor quality of 3D abdominal fake images generated by utilizing random noise as input, deep convolutional neural networks (DCNN) based on feature restoration method is designed to generate more realistic fake images. By testing the proposed algorithm on the LiTS-2017 and KiTS19 dataset, experimental results show that the proposed semi-supervised 3D liver segmentation method can greatly improve the segmentation performance of liver, with a Dice score of 0.9424 outperforming other methods.

M2UNet++

Liver segmentation is instrumental for decision making in the medical realm for the diagnosis and treatment planning of hepatic diseases. However, the manual segmentation of the hundreds of CT images is tedious for medical experts. Thus, it hampers the segmentation accuracy and is reliant on opinion of the operator. This chapter presents the deep learning-based modified multi-scale UNet++ (M2UNet++) approach for automatic liver segmentation. The multi-scale features were modified channel-wise using adaptive feature recalibration to improve the representation of the high-level semantic information of the skip pathways and improved the segmentation performance with fewer computational overheads. The experimental results proved the model's efficacy on the publicly available 3DIRCADb dataset, which offers significant complexity and variations. The model's dice coefficient value is 97.28% that is 7.64%, and 2.24% improved from the UNet and UNet++ model. The quantitative result analysis shows that the M2UNet++ model outperforms the state-of-the-art methods proposed for liver segmentation.

Gabor Dictionary of Sparse Image Patches Selected in Prior Boundaries for 3D Liver Segmentation in CT Images

The gray contrast between the liver and other soft tissues is low, and the boundary is not obvious. As a result, it is still a challenging task to accurately segment the liver from CT images. In recent years, methods of machine learning have become a research hotspot in the field of medical image segmentation because they can effectively use the “gold standard” personalized features of the liver from different data. However, machine learning usually requires a large number of data samples to train the model and improve the accuracy of medical image segmentation. This paper proposed a method for liver segmentation based on the Gabor dictionary of sparse image blocks with prior boundaries. This method reduced the number of samples by selecting the test sample set within the initial boundary area of the liver. The Gabor feature was extracted and the query dictionary was created, and the sparse coefficient was calculated to obtain the boundary information of the liver. By optimizing the reconstruction error and filling holes, a smooth liver boundary was obtained. The proposed method was tested on the MICCAI 2007 dataset and ISBI2017 dataset, and five measures were used to evaluate the results. The proposed method was compared with methods for liver segmentation proposed in recent years. The experimental results show that this method can improve the accuracy of liver segmentation and effectively repair the discontinuity and local overlap of segmentation results.

Improved performance and consistency of deep learning 3D liver segmentation with heterogeneous cancer stages in magnetic resonance imaging

Purpose Accurate liver segmentation is key for volumetry assessment to guide treatment decisions. Moreover, it is an important pre-processing step for cancer detection algorithms. Liver segmentation can be especially challenging in patients with cancer-related tissue changes and shape deformation. The aim of this study was to assess the ability of state-of-the-art deep learning 3D liver segmentation algorithms to generalize across all different Barcelona Clinic Liver Cancer (BCLC) liver cancer stages. Methods This retrospective study, included patients from an institutional database that had arterial-phase T1-weighted magnetic resonance images with corresponding manual liver segmentations. The data was split into 70/15/15% for training/validation/testing each proportionally equal across BCLC stages. Two 3D convolutional neural networks were trained using identical U-net-derived architectures with equal sized training datasets: one spanning all BCLC stages (“All-Stage-Net": AS-Net), and one limited to early and intermediate BCLC stages (“Early-Intermediate-Stage-Net": EIS-Net). Segmentation accuracy was evaluated by the Dice Similarity Coefficient (DSC) on a dataset spanning all BCLC stages and a Wilcoxon signed-rank test was used for pairwise comparisons. Results 219 subjects met the inclusion criteria (170 males, 49 females, 62.8±9.1 years) from all BCLC stages. Both networks were trained using 129 subjects: AS-Net training comprised 19, 74, 18, 8, and 10 BCLC 0, A, B, C, and D patients, respectively; EIS-Net training comprised 21, 86, and 22 BCLC 0, A, and B patients, respectively. DSCs (mean±SD) were 0.954±0.018 and 0.946±0.032 for AS-Net and EIS-Net (p<0.001), respectively. The AS-Net 0.956±0.014 significantly outperformed the EIS-Net 0.941±0.038 on advanced BCLC stages (p<0.001) and yielded similarly good segmentation performance on early and intermediate stages (AS-Net: 0.952±0.021; EIS-Net: 0.949±0.027; p = 0.107). Conclusion To ensure robust segmentation performance across cancer stages that is independent of liver shape deformation and tumor burden, it is critical to train deep learning models on heterogeneous imaging data spanning all BCLC stages.

U-Net combined with multi-scale attention mechanism for liver segmentation in CT images

Background The liver is an important organ that undertakes the metabolic function of the human body. Liver cancer has become one of the cancers with the highest mortality. In clinic, it is an important work to extract the liver region accurately before the diagnosis and treatment of liver lesions. However, manual liver segmentation is a time-consuming and boring process. Not only that, but the segmentation results usually varies from person to person due to different work experience. In order to assist in clinical automatic liver segmentation, this paper proposes a U-shaped network with multi-scale attention mechanism for liver organ segmentation in CT images, which is called MSA-UNet. Our method makes a new design of U-Net encoder, decoder, skip connection, and context transition structure. These structures greatly enhance the feature extraction ability of encoder and the efficiency of decoder to recover spatial location information. We have designed many experiments on publicly available datasets to show the effectiveness of MSA-UNet. Compared with some other advanced segmentation methods, MSA-UNet finally achieved the best segmentation effect, reaching 98.00% dice similarity coefficient (DSC) and 96.08% intersection over union (IOU).