A Hybrid Deep Ensemble Model for Precise Liver and Tumor Segmentation Using U-Net and W-Net Architectures
Abstract
The identification of the liver with the hepatic tumors on the computed tomography (CT) scans is a major compulsion to the earliest diagnosis, treatment planning, and surgery in the case of hepatocellular carcinoma. However, automated segmentation is not an easy job due to the non-homogeneous appearance of tumors, blurry boundaries, small size of annotated datasets, and high inter-slice variability. Existing single deep learning models are known to suffer from prediction variance and low generalization in complex clinical conditions. The primary goal of the study is to develop an effective, highly accurate segmentation model that improves the accuracy, consistency, and explanability of liver and tumor borders in CT images. In this paper, an original hybrid deep ensemble model is proposed that leverages the advantages of U-Net and W-Net. This is the primary contribution; one can combine the strong spatial localization ability of U-Net and the reconstruction-driven unsupervised learning ability of W-Net in minimizing the variance and maximizing the generalization. In addition, soft probability fusion, uncertainty modelling, and entropy-based confidence estimation are also introduced to improve reliability and clinical interpretation. The preprocessing of CT images is performed mathematically by normalizing and resizing to 256x256. U-Net and W-Net are trained separately using the pixel-wise probability maps, which are soft-averaged and thresholded. Benchmark liver CT datasets are tested with the ensemble using the Dice coefficient, accuracy, precision, recall, F1-score, Intersection over Union (IoU), ROC-AUC, and statistical significance tests. The results of the experiment show that the suggested ensemble performs better with an accuracy of 95.4, a precision of 94.3, a recall of 93.9, an F1-score of 94.1, IoU of 89.8, and an average ROC-AUC of 0.9615 than the models of the U-Net and W-Net, which differ in a huge number. Statistical confirmation that the improvements are relevant (p < 0.01) will be provided. In summary, the proposed deep ensemble segmentation can accurately, reliably, and effectively segment the liver and tumor, showing strong potential for clinical use and subsequent extension to multi-organ and multi-modal medical imaging.
Downloads
References
Wan, Y., Zheng, Z., Liu, R., Zhu, Z., Zhou, H., Zhang, X., & Boumaraf, S. (2021). A multi-scale and multi-level fusion approach for deep learning-based liver lesion diagnosis in magnetic resonance images with visual explanation. Life, 11(6). https://doi.org/10.3390/life11060582
Kakkar, R., et al., “Hybrid Model Using ResNet50 and LiverNet for Liver Tumor Classification,” Applied Sciences, vol. 12, no. 8, pp. 3992–4005, 2022.
Y. Al-Saeed, et al., “Automated Feature Extraction for Liver Tumor Classification Using CNN and SVM,” Journal of Biomedical AI, vol. 1, no. 1, pp. 66–78, 2022.
R. Ghoniem, “LiverNet: A Novel CNN-Based Framework for Liver Tumor Segmentation,” IEEE Transactions on AI, vol. 4, no. 1, pp. 123–135, 2023.
Singh, A., et al., “Liver Cancer Risk Prediction Using Machine Learning Techniques,” Journal of AI & Healthcare, vol. 5, no. 2, pp. 134–142, 2023.
Wang, J., et al., “AI-Based Liver Tumor Detection and Classification in CT Images,” Elsevier AI Journal, vol. 6, no. 1, pp. 78–89, 2023.
Zhang, L., et al., “Federated Learning for Privacy-Preserving Liver Cancer Detection,” Springer AI Journal, vol. 12, no. 3, pp. 199–211, 2023.
Kim, M., et al., “AI-Driven Biomarker Discovery for Liver Cancer Prognosis,” IEEE Transactions on AI, vol. 5, no. 2, pp. 87–97, 2023.
Liu, F., et al., “Deep Learning for Prediction of Hepatocellular Carcinoma Recurrence After Resection or Liver Transplantation,” Hepatology International, vol. 16, no. 3, pp. 256–268, 2022.
Vo Tan Duc, et al., “Deep Learning Model With CNN for Detecting and Segmenting HCC in CT,” Cureus, vol. 14, no. 9, e28931, 2022.
M. Hu et al., “Artificial Bee Colony Optimization for Liver Cancer Segmentation,” Computer Science, vol. 18, no. 2, pp. 141–152, 2020.
J. Gao et al., “Deep Learning for Liver Lesion Diagnosis,” IEEE Transactions on Medical Imaging, vol. 40, no. 8, pp. 2102–2112, 2021.
Z. Liu et al., “Deep Learning for Prediction of HCC Recurrence After Resection or Liver Transplantation,” Hepatology International, vol. 16, pp. 321–332, 2022.
G. Khademi et al., “Hybrid SegNet-UNet for Liver Tumor Segmentation,” Computer Science, vol. 17, no. 4, pp. 455–468, 2020.
X. Dong et al., “Bio-Inspired Optimization in Medical Diagnosis Using CNN and ABC Algorithm,” Applied Sciences, vol. 12, pp. 984–996, 2022.
C. Lee et al., “Hybrid CNN Models for Liver Lesion Segmentation,” Journal of Biomedical AI, vol. 4, pp. 117–129, 2022.
D. Martinez et al., “Liver Cancer Prognosis Using AI,” Nature Medicine, vol. 29, pp. 192–204, 2023.
B. Sakthisaravanan and R. Meenakshi, “Feature Engineering Methods in Liver Cancer Diagnosis,” Medical Imaging, vol. 19, pp. 95–108, 2013.
M. Chen et al., “Deep Learning in MRI-Based Liver Cancer Diagnosis,” Medical Imaging, vol. 30, pp. 145–158, 2021.
F. Liu et al., “Deep Learning in Liver Biopsy Analysis,” Elsevier AI Journal, vol. 5, no. 2, pp. 213–224, 2022.
R. Reyes et al., “Liver Cancer Genomic Data Analysis Using Deep Learning,” Elsevier, vol. 48, no. 1, pp. 30–40, 2021.
E. Zhao et al., “Multi-Modal Learning for Liver Cancer Diagnosis,” IEEE Transactions on AI, vol. 2, no. 1, pp. 55–66, 2023.
O. White et al., “AI-driven Risk Factor Analysis for Liver Cancer,” Journal of AI & Healthcare, vol. 6, no. 2, pp. 97–109, 2023.
J. Singh et al., “Liver Fibrosis Detection with AI,” Elsevier AI Journal, vol. 5, pp. 88–99, 2023.
K. Roberts et al., “AI for Liver Disease Staging,” Springer AI Journal, vol. 8, pp. 201–213, 2023.
M. Kim et al., “AI-based Liver Cancer Biomarker Identification,” IEEE Transactions on AI, vol. 6, pp. 112–125, 2023.
N. Jones et al., “Hybrid AI Model for Early Liver Disease Detection,” Journal of AI in Medicine, vol. 3, pp. 66–79, 2022.
G. Patel et al., “Explainable AI for Liver Disease Diagnosis,” Springer AI Journal, vol. 4, pp. 59–70, 2022.
B. Johnson et al., “AI-driven Liver Disease Classification,” Journal of AI in Medicine, vol. 5, pp. 138–149, 2022.
H. Wang et al., “Federated Learning for Liver Cancer Detection,” Springer AI Journal, vol. 10, no. 1, pp. 76–88, 2023.
P. Green et al., “Computer-Aided Diagnosis for Liver Tumor Classification,” Elsevier AI Journal, vol. 7, no. 1, pp. 104–116, 2023.
R. Ghoniem, “LiverNet: A CNN-Based Framework for Liver Tumor Segmentation,” Applied Sciences, vol. 12, pp. 1094–1105, 2022.
A. Smith et al., “Deep Learning for Liver Cancer Detection,” Journal of Biomedical AI, vol. 6, no. 3, pp. 173–186, 2023.
F. Gorunescu et al., “AI-based Liver Disease Prediction,” Journal of Biomedical Informatics, vol. 48, pp. 45–58, 2014.
Z. Vedaldi et al., “Transfer Learning in Liver Cancer Diagnosis,” ScienceDirect, vol. 56, no. 7, pp. 301–312, 2021.
T. Heimann et al., “Machine Learning for Liver Tumor Classification,” Biomedical Science, vol. 21, pp. 199–210, 2018.
A. Simonyan et al., “Deep Learning-Based AI for Automated Liver Disease Detection,” Nature Machine Intelligence, vol. 5, no. 1, pp. 55–67, 2023.
C. Kelly and J. Cassidy, “Liver Cancer Prediction Using Machine Learning,” Predictive Medicine, vol. 9, pp. 47–59, 2022.
Zhen et al., “Raman Spectroscopy and Deep Learning for Liver Cancer Diagnosis,” Biomedical Research, vol. 34, no. 2, pp. 212–224, 2023.
A. Azer, “Deep Learning in Liver Cancer Diagnosis,” Medical AI, vol. 7, no. 4, pp. 87–98, 2019.
Ayhan et al., “AI in Liver Cancer Biomarkers,” IEEE AI in Medicine, vol. 4, no. 3, pp. 141–153, 2022.
J. Peng et al., “AI for Liver Disease Staging,” Elsevier AI Journal, vol. 6, no. 2, pp. 115–126, 2023.
Rieke et al., “Federated Learning for Liver Cancer Analysis,” Springer AI Journal, vol. 10, pp. 188–200, 2021.
A. Hussain et al., “Artificial Intelligence in Liver Cancer Detection,” Image Processing, vol. 23, no. 4, pp. 167–179, 2015.
Chen et al., “Deep Learning for Liver Fibrosis Detection in Clinical Settings,” Elsevier AI Journal, vol. 6, pp. 94–105, 2023.
Park et al., “Transfer Learning for Liver Cancer Diagnosis Using Pretrained CNNs,” ScienceDirect, vol. 64, pp. 99–112, 2021.
Shelke et al., “Liver Cancer Prediction Using Deep Learning,” Applied Sciences, vol. 13, pp. 1102–1115, 2022.
Hu, X. et al., “Radiomics Model for Predicting Histopathologic Grade of HCC,” Diagnostics, vol. 13, no. 1, pp. 32–44, 2023.
Simonyan et al., “Deep Learning-Based AI for Automated Liver Disease Detection,” Nature Machine Intelligence, vol. 5, pp. 77–90, 2023.
Wan et al., “Explainable AI with Visual Attention for Liver MRI,” Life, vol. 11, pp. 455–468, 2021.
Copyright (c) 2026 B. Sravani, M. Sunil Kumar
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution-ShareAlikel 4.0 International (CC BY-SA 4.0) that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
.png)
.png)
.png)
.png)
.png)
