Hybrid Deep Learning for CT-Based Liver Lesion Classification

International Journal of Electronics and Communication Engineering

© 2026 by SSRG - IJECE Journal

Volume 13 Issue 3

Year of Publication : 2026

Authors : Trupti M. Kodinariya, Nikhil Gondaliya

10.14445/23488549/IJECE-V13I3P110

How to Cite?

Trupti M. Kodinariya, Nikhil Gondaliya, "Hybrid Deep Learning for CT-Based Liver Lesion Classification," SSRG International Journal of Electronics and Communication Engineering, vol. 13,  no. 3, pp. 129-141, 2026. Crossref, https://doi.org/10.14445/23488549/IJECE-V13I3P110

Abstract:

Traditional visual evaluation is subjective and differs among observers, making an accurate identification of liver lesions on abdominal CT scans crucial for the prompt planning of treatment. In order to classify liver lesions as benign or malignant, this article presents a segmentation-free hybrid deep learning method that uses the LiTS-2017 dataset. The method involves combining systematic pre-processing with parallel feature extraction using Xception and Swin Transformer networks for synergistic local-global representations. Various fusion methodologies were evaluated, and a gated feature fusion mechanism was implemented to integrate distinct blocks utilizing channel and spatial attention modules, thereby adaptively weighting and recalibrating the salient characteristics prior to classification. This approach outperforms all other methods of model fusion and individual models in terms of accuracy (99.93%), recall (99.92%), F1 score (99.90%), and classification accuracy (99.93%), according to the experimental data. By eliminating the necessity for segmentation, our suggested architecture shows great promise as a trustworthy clinical decision support tool for differentiating between benign and malignant liver lesions using adaptive feature fusion.

Keywords:

Abdominal CT imaging, Hybrid Deep Learning, LiTS17 Dataset, Liver lesion classification, Segmentation-free classification, Swin Transformer, Xception.

References:

[1] Jiarong Zhou et al., “Automatic Detection and Classification of Focal Liver Lesions based on Deep Convolutional Neural Networks: A Preliminary Study,” Frontiers in Oncology, vol. 10, pp. 1-11, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[2] Francisco Perdigón Romero et al., “End-To-End Discriminative Deep Network for Liver Lesion Classification,” 2019 IEEE 16th International Symposium on Biomedical Imaging (ISBI 2019), Venice, Italy, pp. 1243-4246, 2019.
[CrossRef] [Google Scholar] [Publisher Link]
[3] Maayan Frid-Adar et al., “GAN-based Synthetic Medical Image Augmentation for Increased CNN Performance in Liver Lesion Classification,” Neurocomputing, vol. 321, pp. 321-331, 2018.
[CrossRef] [Google Scholar] [Publisher Link]
[4] Mubasher Hussain, Najia Saher, and Salman Qadri, “Computer Vision Approach for Liver Tumor Classification Using CT Dataset,” Applied Artificial Intelligence, vol. 36, no. 1, pp. 1-24, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[5] Ayesha Amir Siddiqi, Attaullah Khawaja, and Adnan Hashmi, “Classification of Abdominal CT Images bearing Liver Tumor Using Structural Similarity Index and Support Vector Machine,” Mehran University Research Journal of Engineering and Technology, vol. 39, no. 4, pp. 751-758, 2020.
[CrossRef] [Google Scholar] [Publisher Link]
[6] Su-E Cao et al., “Multiphase Convolutional Dense Network for the Classification of Focal Liver Lesions on Dynamic Contrast-Enhanced Computed Tomography,” World Journal of Gastroenterology, vol. 26, no. 25, pp. 3660-3672, 2020.
[CrossRef] [Google Scholar] [Publisher Link]
[7] Yuankai Huo et al., “Harvesting, Detecting, and Characterizing Liver Lesions from Large-scale Multi-phase CT Data via Deep Dynamic Texture Learning,” arxiv Preprint, pp. 1-10, 2020.
[CrossRef] [Google Scholar] [Publisher Link]
[8] Sang-Gil Lee et al., “Robust End-to-End Focal Liver Lesion Detection Using Unregistered Multiphase Computed Tomography Images,” IEEE Access, vol. 7, no. 2, pp. 319-329, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[9] Ling Zhao et al., “A Unified End-to-End Classification Model for Focal Liver Lesions,” Biomedical Signal Processing and Control, vol. 86, 2023.
[CrossRef] [Google Scholar] [Publisher Link]
[10] Lei Wu et al., “Beyond Radiologist-level Liver Lesion Detection on Multi-Phase Contrast-Enhanced CT Images by Deep Learning,” iScience, vol. 26, no. 11, pp. 1-17, 2023.
[CrossRef] [Google Scholar] [Publisher Link]
[11] Zhehan Shen et al., “An Explainable Deep Learning Model for Focal Liver Lesion Diagnosis Using Multiparametric MRI,” Radiology: Artificial Intelligence, vol. 7, no. 6, 2025.
[CrossRef] [Google Scholar] [Publisher Link]
[12] Robert Stollmayer et al., “Diagnosis of Focal Liver Lesions with Deep Learning-based Multi-channel Analysis of Hepatocyte-Specific Contrast-Enhanced Magnetic Resonance Imaging,” World Journal of Gastroenterology, vol. 27, no. 35, pp. 5978-5988, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[13] Anh-Cang Phan et al., “Improving Liver Lesions Classification on CT/MRI Images Based on Hounsfield Units Attenuation and Deep Learning,” Gene Expression Patterns, vol. 47, 2023.
[CrossRef] [Google Scholar] [Publisher Link]
[14] Hansang Lee et al., “Classification of Focal Liver Lesions in CT Images using Convolutional Neural Networks with Lesion Information Augmented Patches and Synthetic Data Augmentation,” Medical Physics, vol. 48, no. 9, pp. 5029-5046, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[15] Amitojdeep Singh, Sourya Sengupta, and Vasudevan Lakshminarayanan, “Explainable Deep Learning Models in Medical Image Analysis,” Journal of Imaging, vol. 6, no. 6, pp. 1-19, 2020.
[CrossRef] [Google Scholar] [Publisher Link]
[16] Fei Lyu et al., “Weakly Supervised Liver Tumor Segmentation using Couinaud Segment Annotation,” IEEE Transactions on Medical Imaging, vol. 41, no. 5, pp. 1138-1149, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[17] Chiu Sung-Hua et al., “Binary Classification of Benign and Malignant Hepatic Lesions with Portal Venous Phase Computed Tomography Images with Deep Learning: A Single-institution Study,” Journal of Medical Sciences, vol. 45, no. 2, pp. 33-37, 2025.
[CrossRef] [Google Scholar] [Publisher Link]
[18] Shaohua Qiao et al., “Four-phase CT Lesion Recognition Based on Multi-phase Information Fusion Framework and Spatiotemporal Prediction Module,” BioMedical Engineering OnLine, vol. 23, pp. 1-18, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[19] Patrick Bilic et al., “The Liver Tumor Segmentation Benchmark (LiTS),” Medical Image Analysis, vol. 84, pp. 1-24, 2023.
[CrossRef] [Google Scholar] [Publisher Link]
[20] Hyuna Sung et al., “Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries,” CA: A Cancer Journal for Clinicians, vol. 71, no. 3, pp. 209-249, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[21] Alejandro Forner, María Reig, and Jordi Bruix, “Hepatocellular Carcinoma,” The Lancet, vol. 391, pp. 1301-1314, 2018.
[CrossRef] [Google Scholar] [Publisher Link]
[22] Khaled Y. Elbanna, and Ania Z. Kielar, “Computed Tomography Versus Magnetic Resonance Imaging for Hepatic Lesion Characterization/ Diagnosis,” Clinical Liver Disease, vol. 17, no. 3, 159-164, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[23] Wenya Linda Bi et al., “Artificial Intelligence in Cancer Imaging: Clinical Challenges and Applications,” CA: A Cancer Journal for Clinicians, vol. 69, no. 2, pp. 127-157, 2019.
[CrossRef] [Google Scholar] [Publisher Link]
[24] Koichiro Yasaka, and Osamu Abe, “Deep Learning and Artificial Intelligence in Radiology: Current Applications and Future Directions,” PLOS Medicine, vol. 15, no. 11, pp. 1-14, 2018.
[CrossRef] [Google Scholar] [Publisher Link]
[25] Hayit Greenspan, Bram van Ginneken, and Ronald M. Summers, “Guest Editorial Deep Learning in Medical Imaging: Overview and Future Promise of an Exciting New Technique,” IEEE Transactions on Medical Imaging, vol. 35, no. 5, pp. 1153-1159, 2016.
[CrossRef] [Google Scholar] [Publisher Link]
[26] Geert Litjens et al., “A Survey on Deep Learning in Medical Image Analysis,” Medical Image Analysis, vol. 42, pp. 60-88, 2017.
[CrossRef] [Google Scholar] [Publisher Link]
[27] Patrick Ferdinand Christ et al., “Automatic Liver and Tumor Segmentation of CT and MRI Volumes using Cascaded Fully Convolutional Neural Networks,” arXiv Preprint, pp. 1-20, 2017.
[CrossRef] [Google Scholar] [Publisher Link]
[28] Kaiming He et al., “Deep Residual Learning for Image Recognition,” Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, USA, pp. 770-77, 2016.
[CrossRef] [Google Scholar] [Publisher Link]
[29] Christian Szegedy et al., “Inception-v4, Inception-Resnet and the Impact of Residual Connections on Learning,” Thirty-First Proceedings of the AAAI Conference on Artificial Intelligence, vol. 31, no. 1, pp. 4278-4284, 2017.
[CrossRef] [Google Scholar] [Publisher Link]
[30] Khaled Alawneh et al., “LiverNet: Diagnosis of Liver Tumors in Human CT Images,” Applied Sciences, vol. 12, no. 11, pp. 1-16, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[31] Shivani Joshi et al., “Enhancing Liver Cancer Detection: An Innovative Deep Learning Approach Combining GAN, ResNet, and Vision Transformer,” Expert Systems with Applications, vol. 298, 2026.
[CrossRef] [Google Scholar] [Publisher Link]
[32] R. Archana, and L. Anand, “Multi-Scale Cascaded Spatial Segmentation Transformer for Liver Cancer Classification,” International Journal of Computational Intelligence Systems, pp. 1-32, 2025.
[CrossRef] [Google Scholar] [Publisher Link]
[33] Guang Mei et al., “Research on CT Image Segmentation and Classification of Liver Tumors based on Attention Mechanism and Improved U-Net Model,” Technology and Health Care, vol. 33, no. 5, pp. 2468-2483, 2025.
[CrossRef] [Google Scholar] [Publisher Link]
[34] Srinivas Kolli et al., “A Novel Liver Tumor Classification using Improved Probabilistic Neural Networks with Bayesian Optimization,” e-Prime - Advances in Electrical Engineering, Electronics and Energy, vol. 8, pp. 1-9, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[35] P.R. Aparna, and T.M. Libish, “Automatic Segmentation and Classification of the Liver Tumor using Deep Learning Algorithms,” 2023 3rd International Conference on Advances in Computing, Communication, Embedded and Secure Systems (ACCESS), Kalady, Ernakulam, India, pp. 334-339, 2023.
[CrossRef] [Google Scholar] [Publisher Link] 
[36] François Chollet, “Xception: Deep Learning with Depthwise Separable Convolutions,” 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, HI, USA, pp. 1800-1807, 2017.
[CrossRef] [Google Scholar] [Publisher Link]
[37] Reza Azad et al., “Advances in Medical Image Analysis with Vision Transformers: A Comprehensive Review,” Medical Image Analysis, vol. 91, pp. 1-72, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[38] Xiayu Guo et al., “UCTNet: Uncertainty-Guided CNN-Transformer Hybrid Networks for Medical Image Segmentation,” Pattern Recognition, vol. 152, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[39] Aymen M. Al-Hejri et al., “A Hybrid Vision Transformer with Ensemble CNN Framework for Cervical Cancer Diagnosis,” BMC Medical Informatics and Decision Making, vol. 25, pp. 1-20, 2025.
[CrossRef] [Google Scholar] [Publisher Link]
[40] Xiaolei He et al., “A Novel Liver Image Classification Network for Accurate Diagnosis of Liver Diseases,” Scientific Reports, vol. 15, pp. 1-16, 2025.
[CrossRef] [Google Scholar] [Publisher Link]