An Attention-Enhanced U-Net Model for Precise Liver Tumor Segmentation Using Convolutional Block Attention and Efficient Neural Network: Advancing Automated Imaging for Improved Liver Tumor Diagnosis

International Journal of Electronics and Communication Engineering

© 2025 by SSRG - IJECE Journal

Volume 12 Issue 12

Year of Publication : 2025

Authors : B. Shashikanth, K. Sivani

10.14445/23488549/IJECE-V12I12P105

How to Cite?

B. Shashikanth, K. Sivani, "An Attention-Enhanced U-Net Model for Precise Liver Tumor Segmentation Using Convolutional Block Attention and Efficient Neural Network: Advancing Automated Imaging for Improved Liver Tumor Diagnosis," SSRG International Journal of Electronics and Communication Engineering, vol. 12,  no. 12, pp. 44-59, 2025. Crossref, https://doi.org/10.14445/23488549/IJECE-V12I12P105

Abstract:

Liver Tumor Segmentation in CT images is an essential step in the medical imaging process, which is generally difficult due to the variability of tumor shapes, sizes, and ambiguous boundaries with healthy liver tissue. This paper proposes a new model that integrates the U-Net structure and Convolutional Block Attention Module along with EfficientNet-B0 to promote both segmentation performance and computational efficiency. Our proposed model is distinctive because it incorporates CBAM to recalibrate the spatial and channel-wise attention maps adaptively, focusing on the informative tumor regions for improved segmentation performance. EfficientNet-B0, an encoder backbone characterized by the compound scaling method and lightweight structure, is employed to enhance hierarchical feature extraction and computational efficiency. The model achieved a mean Intersection over Union (IoU) of 0.9356 on a public dataset. This result significantly outperforms strong previous methods, such as PSP (0.9113) and MANet (0.6555). Our findings show that the incorporation of attention modules with lightweight encoders is effective for precise liver tumor segmentation, and the proposed method has high potential for clinical applications. This paper paves the way for the potential of innovative and scalable diagnostics in the field of health care.

Keywords:

Liver Tumor Segmentation, U-Net, CBAM, EfficientNet-B0, Medical Imaging, IoU.

References:

[1] Ju Zou et al., “Neuroimmune Modulation in Liver Pathophysiology,” Journal of Neuroinflammation, vol. 21, no. 1, pp. 1-25, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[2] Sohini Goswami et al., “The Alarming Link between Environmental Microplastics and Health Hazards with Special Emphasis on Cancer,” Life Sciences, vol. 355, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[3] Barsha Abhisheka et al., “Recent Trend in Medical Imaging Modalities and their Applications in Disease Diagnosis: a Review,” Multimedia Tools and Applications, vol. 83, pp. 43035-43070, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[4] Haoyang Jiang et al., “Deep Learning for Liver Cancer Histopathology Image Analysis: A Comprehensive Survey,” Engineering Applications of Artificial Intelligence, vol. 133, pp. 1-22, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[5] Ravi Rai Dangi, Anil Sharma, and Vipin Vageriya, “Transforming Healthcare in Low‐Resource Settings with Artificial Intelligence: Recent Developments and Outcomes,” Public Health Nursing, vol. 42, no. 2, pp. 1017-1030, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[6] R. Archana, and P.S. Eliahim Jeevaraj, “Deep Learning Models for Digital Image Processing: A Review,” Artificial Intelligence Review, vol. 57, pp. 1-33, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[7] Jean-Charles Nault, Julien Calderaro, and Maxime Ronot, “Integration of New Technologies in the Multidisciplinary Approach to Primary Liver Tumours: The Next-Generation Tumour Board,” Journal of Hepatology, vol. 81, no. 4, pp. 756-762, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[8] Chaopeng Wu et al., “A Review of Deep Learning Approaches for Multimodal Image Segmentation of Liver Cancer,” Journal of Applied Clinical Medical Physics, vol. 25, no. 12, pp. 1-22, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[9] Chuanfei Hu et al., “Trustworthy Multi-phase Liver Tumor Segmentation via Evidence-based Uncertainty,” Engineering Applications of Artificial Intelligence, vol. 133, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[10] S. Saumiya, and S. Wilfred Franklin, “Unified Automated Deep Learning Framework for Segmentation and Classification of Liver Tumors,” The Journal of Supercomputing, vol. 80, pp. 2347-2380, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[11] Arifullah et al., For the Nuclei Segmentation of Liver Cancer Histopathology Images, A Deep Learning Detection Approach is Used, Engineering Applications of Artificial Intelligence, Springer, Cham, pp. 263-274, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[12] Yuanyuan Shui et al., “A Three-Path Network with Multi-scale Selective Feature Fusion, Edge-Inspiring and Edge-Guiding for Liver Tumor Segmentation,” Computers in Biology and Medicine, vol. 168, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[13] Yilin You et al., “Contour-Induced Parallel Graph Reasoning for Liver Tumor Segmentation,” Biomedical Signal Processing and Control, vol. 92, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[14] Raju Egala, and M.V.S. Sairam, “A Review on Medical Image Analysis Using Deep Learning,” Engineering Proceedings, vol. 66, no. 1, pp. 1-4, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[15] Kai-Ni Wang et al., “SBCNet: Scale and Boundary Context Attention Dual-Branch Network for Liver Tumor Segmentation,” IEEE Journal of Biomedical and Health Informatics, vol. 28, no. 5, pp. 2854-2865, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[16] Ali Mohammed Hend et al., “Adaptive Method for Exploring Deep Learning Techniques for Subtyping and Prediction of Liver Disease,” Applied Sciences, vol. 14, no. 4, pp. 1-19, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[17] Jinke Wang et al., “EAR-U-Net: EfficientNet and Attention-Based Residual U-Net for Automatic Liver Segmentation in CT,” arXiv preprint, pp. 1-26, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[18] Benyue Zhang, Shi Qiu, and Ting Liang, “Dual Attention-Based 3D U-Net Liver Segmentation Algorithm on CT Images,” Bioengineering, vol. 11, no. 7, pp. 1-20, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[19] Wanrat Limprapaipong et al., “A Paired Multi-Scale Attention Network for Liver Tumor Segmentation in 99mTc-MAA SPECT/CT Imaging,” Scientific Reports, vol. 15, pp. 1-16, 2025.
[CrossRef] [Google Scholar] [Publisher Link]
[20] T. Robin, Data Unet, Kaggle, 2023. [Online]. https://www.kaggle.com/datasets/robintrmbtt/data-unet
[21] Mengke Ma et al., “A Novel Model for Predicting Postoperative Liver Metastasis in R0 Resected Pancreatic Neuroendocrine Tumors: Integrating Computational Pathology and Deep Learning-Radiomics,” Journal of Translational Medicine, vol. 22, pp. 1-14, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[22] Sukanya Saeku et al., “Liver and Tumor Segmentation in Selective Internal Radiation Therapy 99mTc-MAA SPECT/CT Images using MANet and Histogram Adjustment,” 2022 3rd Asia Symposium on Signal Processing (ASSP), pp. 62-66, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[23] Jia-Qi Zhu et al., “Responsive Hydrogels Based on Triggered Click Reactions for Liver Cancer,” Advanced Materials, vol. 34, no. 38, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[24] Jaspreet Kaur, and Prabhpreet Kaur, “PSO-PSP-Net + InceptionV3: An Optimized Hyper-Parameter Tuned Computer-Aided Diagnostic Model for Liver Tumor Detection using CT Scan Slices,” Biomedical Signal Processing and Control, vol. 95, 2024.
[CrossRef] [Google Scholar] [Publisher Link]