Xin Qiao
ABSTRACT
The accurate segmentation of liver tumors is the premise of judging the scope of liver tumors and formulating treatment plan. However, the traditional methods of liver tumor segmentation have the problems of high complexity and low precision. In order to address these issues, we propose a novel semantic localization and centroid region growth approach in this work. Firstly, the 3D U-Net+ deep learning model is adopted to extract features and locate the liver. Secondly, appropriate initial seed points are selected in the liver by the centroid region growth algorithm, and the tumor regions are gradually grown and segmented according to the growth criteria. To enhance the precision of segmentation even further, post-processing steps are introduced to remove the holes in the tumor region and perform boundary smoothing. The experimental outcomes demonstrate that the algorithm presented attains exceptional segmentation precision and robustness when applied to the LITS2017 dataset. Therefore, the algorithm has the potential to be used as a reliable tool in clinical practice to provide physicians with accurate liver tumor localization and segmentation results.
- Xia Y, Xie X, Wu X, An Approach of Automatically Selecting Seed Point Based on Region Growing for Liver Segmentation [C]//2019 8th International Symposium on Next Generation Electronics (ISNE). 2019. DOI:10. 1109/ISNE. 2019. 8896442.Google Scholar
- Kolařík, Martin, Burget R, Uher, Václav, Optimized High Resolution 3D Dense-U-Net Network for Brain and Spine Segmentation [J]. Applied Sciences, 2019, 9(3). DOI:10. 3390/app9030404.Google Scholar
- Hong Y, Mao X W, Hui Q L, Automatic liver and tumor segmentation based on deep learning and globally optimized refinement [J]. Applied Mathematics, 2021, 36(2):304-316. DOI:10. 1007/s11766-021-4376-3.Google ScholarCross Ref
- Wu Y, Liao K, Chen J, D-former: a U-shaped Dilated Transformer for 3D medical image segmentation [J]. Neural computing & applications, 2023.Google Scholar
- Yang Z H . Liver segmentation based on region growing and level set active contour model with new signed pressure force function [J]. Optik: Zeitschrift fur Licht- und Elektronenoptik: = Journal for Light-and Electronoptic, 2020, 202.Google Scholar
- Yang Z, Zhao Y Q, Liao M, Semi-automatic liver tumor segmentation with adaptive region growing and graph cuts [J]. Biomedical signal processing and control, 2021 (Jul. Pt. 2):68.Google Scholar
- Baazaoui, A, Ahmed, Semi-Automated Segmentation of Single and Multiple Tumors in Liver CT Images Using Entropy-Based Fuzzy Region Growing [J]. Innovation and research in biomedical engineering: IRBM, 2017.Google Scholar
- XIAO M Y, LI X F, ZHANG X L, ZHANG L. Medical image segmentation algorithm based on multi-scale region growing. Journal of Jilin University (Engineering Edition), 2017.Google Scholar
- Biratu E S, Schwenker F, Debelee T G, Enhanced Region Growing for Brain Tumor MR Image Segmentation [J]. Journal of Imaging, 2021, 7(2):22. DOI:10. 3390/jimaging7020022.Google ScholarCross Ref
- Hong Y, Mao X W, Hui Q L, Automatic liver and tumor segmentation based on deep learning and globally optimized refinement [J]. Applied Mathematics, 2021, 36(2):304-316. DOI:10. 1007/s11766-021-4376-3.Google ScholarCross Ref
- Shichao J, Yanjun S, Shang G, Deep Learning: Individual Maize Segmentation From Terrestrial Lidar Data Using Faster R-CNN and Regional Growth Algorithms [J]. Frontiers in Plant Science, 2018, 9:866-. DOI:10. 3389/fpls. 2018. 00866.Google ScholarCross Ref
- Lv P, Wang J, Wang H . 2. 5D lightweight RIU-Net for automatic liver and tumor segmentation from CT [J]. Biomedical signal processing and control, 2022(May):75. DOI:10. 1016/j. bspc. 2022. 103567.Google Scholar
- He F, Zhang G, Yang H, Multi-scale Attention Module U-Net liver tumour segmentation method [J]. Journal of Physics Conference Series, 2020, 1678:012107. DOI:10. 1088/1742-6596/1678/1/012107.Google ScholarCross Ref
- Zhang C, Lu J, Hua Q, SAA-Net: U-shaped network with Scale-Axis-Attention for liver tumor segmentation [J]. Biomedical Signal Processing and Control, 2022, 73:103460-. DOI:10. 1016/j. bspc. 2021. 103460.Google ScholarCross Ref
- Zeng Y Z, Zhao Y Q, Liao M, Liver vessel segmentation based on extreme learning machine [J]. Physica Medica, 2016, 32(5):709-716. DOI:10. 1016/j. ejmp. 2016. 04. 003.Google ScholarCross Ref
- Eugene Vorontsov, An Tang, David Roy, Metastatic liver tumour segmentation with a neural network-guided 3D deformable model [J]. Medical & Biological Engineering & Computing, 2017. DOI:10. 1007/s11517-016-1495-8.Google Scholar
Index Terms
- Accurate liver tumors segmentation using semantic localization and centroid region growth
Recommendations
Automatic liver tumor segmentation from CT images using hierarchical iterative superpixels and local statistical features
AbstractLiver tumor segmentation from CT images plays an important role in disease diagnosis and treatment planning. In this paper, we propose an automatic segmentation framework dedicated to accurate liver tumor extraction from CT images. To ...
Highlights- A local information based SLIC is introduced for superpixel decomposition.
- A ...
Registration-Propagated Liver Tumor Segmentation for Non-enhanced CT-Based Interventions
Image and GraphicsAbstractIntra-operative non-enhanced computed tomography (CT) has grown as a prominent tool in interventional surgeries for liver tumor by offering a real-time, more precise imaging guidance over its pre-operative enhanced counterpart as well as intra-...
Deep learning techniques for liver and liver tumor segmentation: A review
AbstractLiver and liver tumor segmentation from 3D volumetric images has been an active research area in the medical image processing domain for the last few decades. The existence of other organs such as the heart, spleen, stomach, and kidneys ...
Highlights- An extensive review of deep learning architectures for semantic segmentation is provided.
- All published cutting-edge deep learning methodologies for liver segmentation and lesion segmentation are discussed.
- Various challenges for ...
Comments