BCIW-Net: Boundary-Conditioned Adaptive CT Windowing for Hepatocellular Carcinoma Segmentation
5 affiliations
- 1Department of Computer Engineering and Mathematics, Universitat Rovira i Virgili, 43007 Tarragona, Spain
- 2Centre for Cancer Biomarkers & Biotherapeutics, Barts Cancer Institute, Queen Mary University of London, London EC1M 6BQ, UK
- 3Department of Pathology and Clinical Bioinformatics, Erasmus MC, 3015 GD Rotterdam, The Netherlands
- 4Queen’s University Belfast, Belfast BT7 1NN, UK
- 5Electrical Engineering Department, Faculty of Engineering, Aswan University, Aswan 81528, Egypt
Abstract
Background/Objectives: Hepatocellular carcinoma (HCC) is the most common primary liver malignancy, and accurate tumour segmentation in contrast-enhanced computed tomography is essential for treatment planning, assessment of ablation margins, and post-treatment evaluation. Existing deep learning methods typically use fixed-intensity CT preprocessing, overlooking the adaptive windowing employed by radiologists to enhance tumour visibility. To address this limitation, we propose BCIW-Net, a boundary-conditioned intensity-windowed network that integrates spatially adaptive and differentiable intensity windowing into the segmentation process as a clinically motivated inductive bias. Methods: The proposed framework first generates complementary soft-tissue and tumour-enhancing intensity representations using differentiable sigmoid-based transformations. A primary encoder–decoder module predicts tumour guidance, boundary probabilities, and pixel-wise error proxy maps, which collectively guide the dynamic estimation of spatially varying window centre and width parameters. The resulting adaptively windowed features are selectively blended with fixed-intensity representations through an error-gated mechanism before final refinement in a secondary encoder–decoder network. A multi-task objective jointly optimises tumour segmentation, boundary prediction, error-proxy estimation, and adaptive window regularisation in an end-to-end manner. Results: Experiments on the primary liver cancer (PLC)-CECT dataset demonstrate that BCIW-Net achieved a mean Dice similarity coefficient of 78.68% and an intersection over union (IoU) of 64.85% across five-fold patient-level cross-validation, outperforming the seven evaluated CNN-, Transformer-, and state-space baselines. Evaluation on LiTS2017 without dataset-specific fine-tuning yielded a tumour Dice score of 89.84%, providing preliminary evidence of cross-dataset generalisation. Conclusions: Qualitative feature visualisations show that adaptive windowing enhances tumour representation and boundary delineation, demonstrating the effectiveness and interpretability of BCIW-Net for HCC segmentation.
Related
- Convolutional, State-Space and Transformer Liver-Tumor Segmentation Models Trained on Hepatocellular Carcinoma Fail on the Same External Cases: Lesion Size Outweighs Architecture in Cross-Dataset Generalization
- Evaluation of boundary-aware UNet++ for low-grade glioma segmentation
- Boundary-aware multi-scale residual learning with hybrid radiomic–deep fusion for lung tumor segmentation and classification
- AMF-U-Net: an adaptive multimodal fusion residual attention 3D U-Net for boundary-aware brain tumour segmentation
- Preoperative Zinc Deficiency as a Novel Prognostic Predictor for Postoperative Recurrence in Hepatocellular Carcinoma
BibTeX
@article{singh2026bciwnet,
title = {{BCIW-Net: Boundary-Conditioned Adaptive CT Windowing for Hepatocellular Carcinoma Segmentation}},
author = {Tripti Singh and Julia Rebecca Kampel and Farhan Akram and Yifan Li and Akshat Tiwari and Sheetal Santosh Deshpande and Shriya Bommaraveni and Mohamed Abdel-Nasser and Domenec Puig and Vivek Kumar Singh},
journal = {Cancers},
year = {2026},
month = oct,
doi = {10.3390/cancers18203317},
url = {https://www.mdpi.com/2072-6694/18/20/3317},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - BCIW-Net: Boundary-Conditioned Adaptive CT Windowing for Hepatocellular Carcinoma Segmentation
AU - Tripti Singh
AU - Julia Rebecca Kampel
AU - Farhan Akram
AU - Yifan Li
AU - Akshat Tiwari
AU - Sheetal Santosh Deshpande
AU - Shriya Bommaraveni
AU - Mohamed Abdel-Nasser
AU - Domenec Puig
AU - Vivek Kumar Singh
PY - 2026
DA - 2026/10/01
JO - Cancers
DO - 10.3390/cancers18203317
UR - https://www.mdpi.com/2072-6694/18/20/3317
AB - Background/Objectives: Hepatocellular carcinoma (HCC) is the most common primary liver malignancy, and accurate tumour segmentation in contrast-enhanced computed tomography is essential for treatment planning, assessment of ablation margins, and post-treatment evaluation. Existing deep learning methods typically use fixed-intensity CT preprocessing, overlooking the adaptive windowing employed by radiologists to enhance tumour visibility. To address this limitation, we propose BCIW-Net, a boundary-conditioned intensity-windowed network that integrates spatially adaptive and differentiable intensity windowing into the segmentation process as a clinically motivated inductive bias. Methods: The proposed framework first generates complementary soft-tissue and tumour-enhancing intensity representations using differentiable sigmoid-based transformations. A primary encoder–decoder module predicts tumour guidance, boundary probabilities, and pixel-wise error proxy maps, which collectively guide the dynamic estimation of spatially varying window centre and width parameters. The resulting adaptively windowed features are selectively blended with fixed-intensity representations through an error-gated mechanism before final refinement in a secondary encoder–decoder network. A multi-task objective jointly optimises tumour segmentation, boundary prediction, error-proxy estimation, and adaptive window regularisation in an end-to-end manner. Results: Experiments on the primary liver cancer (PLC)-CECT dataset demonstrate that BCIW-Net achieved a mean Dice similarity coefficient of 78.68% and an intersection over union (IoU) of 64.85% across five-fold patient-level cross-validation, outperforming the seven evaluated CNN-, Transformer-, and state-space baselines. Evaluation on LiTS2017 without dataset-specific fine-tuning yielded a tumour Dice score of 89.84%, providing preliminary evidence of cross-dataset generalisation. Conclusions: Qualitative feature visualisations show that adaptive windowing enhances tumour representation and boundary delineation, demonstrating the effectiveness and interpretability of BCIW-Net for HCC segmentation.
SN - 2072-6694
ER -