Jul 2026· IEEE International Conference on Consumer Electronics· pp. 691-696· 0 citations· 11 references
Abstract
Gastrointestinal (GI) endoscopy is a cornerstone diagnostic procedure for detecting inflammatory diseases, polyps, and early-stage cancers. Recent advances in deep learning have significantly improved automated endoscopic image analysis; however, their performance remains limited by scarce annotations, severe class imbalance, and poor generalization across diverse imaging conditions. Moreover, jointly learning classification and segmentation poses additional challenges due to task imbalance and the high annotation cost of pixel-level labels. To address these limitations, we propose a Balanced-Mix-Driven framework that leverages 99,417 unlabeled images from the HyperKvasir dataset through Self-Supervised Learning (SSL)-based pretraining. Our core contribution, Balanced-Mix, is an interpolation strategy that progressively shifts from coarse to fine-grained mixing during pretraining, preventing trivial representation learning. Experimental results on the Kvasir-SEG dataset demonstrate that our method achieves a Dice score of $\mathbf{9 2 . 4 0} \boldsymbol{\%}$ and an mIoU of $\mathbf{8 7 . 0 1 \%}$, outperforming established baselines such as UNet++ and ResUNet++. This validates the effectiveness of curriculum-based self-supervised learning for dense medical prediction tasks.
Colorectal polyps are primarily detected through colonoscopy, which plays a central role in early cancer prevention. Precise polyp segmentation supports treatment planning and diagnostic accuracy by providing masks that encode clinically relevant structures. Recent advancements in deep learning have led to several polyp segmentation models. However, performance remains hindered by challenges such as image noise, complex textures, indistinct boundaries, and diverse polyp morphologies. The high cost and time burden of manual annotation underscore the need for automated segmentation systems. To overcome these limitations, BAASNet, a Boundary-Aware Attention-Based Segmentation framework, is introduced for polyp segmentation. A boundary-aware loss function is integrated to improve performance, particularly in delineating polyp edges. The method is evaluated on nine publicly available datasets spanning five imaging modalities, including two center-wise polyp detection benchmarks, demonstrating strong generalization capability. On PolypDB, the model attains a mean Dice similarity coefficient (mDSC) of at least \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$89.60\%$$\end{document} across all five modalities. Across all evaluated benchmarks, the proposed model achieves an average absolute improvement of approximately \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$3.3\%$$\end{document} in Dice. Gains vary by dataset, ranging from approximately \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$0.7\%$$\end{document} to \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$4.7\%$$\end{document} relative improvement over the best previous results. These results demonstrate BAASNet’s potential for robust, real-time clinical deployment in automated colonoscopy workflows.
Khola Naseem, Nabeel Khalid, A. Dengel et al.· Scientific Reports· 0 citations
Accurate differentiation between gastric adenoma and carcinoma during endoscopy is critical for clinical decision-making. Yet, this task is highly challenging due to high inter-class similarity and ambiguous boundaries between the two classes. Existing ROI-based classification methods often suffer from detection/segmentation error propagation and loss of surrounding global context. In contrast, full-image classification lacks the necessary spatial focus. Furthermore, we observe that deep neural networks gravitate towards domain-specific texture biases(e.g. bleeding, lighting artifacts), often causing models to predict based on spurious correlations instead of intrinsic morphological features. To address these limitations, we propose a novel framework, Masked Achromatic Guidance Expert (MAGE). During training, we introduce an auxiliary local expert branch trained on masked achromatic views of the neoplasm. By suppressing background context and color, this branch is forced to learn highly discriminative, purely structural features. We then employ a dual-objective distillation strategy, transferring both classification logits and spatial attention maps to provide implicit spatial supervision to the main branch that receives full WLI as input. This dual-objective distillation forces the model to ground its predictions in morphology rather than relying on shortcuts, while still retaining clinically relevant color cues. At inference time, our deployable model operates on images without annotated masks, ensuring real-time deployability . Extensive experiments on a clinical gastric endoscopy dataset show that our method significantly outperforms existing detection-based methodologies (e.g. YOLO) and classification-based methodologies (e.g. Swin-Transformer), providing not only superior classification performance but also interpretable attention maps for clinical reliability.
Jiho Jun, Jeongwon Woo, Jaemin Song et al.· arXiv.org· 0 citations
Accurate polyp segmentation is essential for early detection of colorectal cancer, where precise delineation of lesion boundaries directly impacts clinical decision-making. Despite significant progress, existing convolutional methods often struggle to capture global contextual information, while transformerbased and foundation models introduce high computational complexity and require extensive fine-tuning. In this work, we propose LS-PolypSeg, a parameter-efficient framework that leverages a pretrained SAM3 vision encoder with Low-Rank Adaptation (LoRA) for domain-specific learning. By selectively adapting key transformer layers while keeping most of the encoder frozen, the proposed approach preserves rich pretrained representations while significantly reducing training overhead. To complement global feature extraction, a lightweight UNetstyle decoder performs multi-scale feature fusion, enabling accurate recovery of fine-grained spatial details. Extensive experiments on three benchmark datasets, namely Kvasir-SEG, CVC-ClinicDB, and BKAI-IGH, demonstrate that LS-PolypSeg achieves competitive and state-of-the-art performance across multiple evaluation metrics. These results highlight the effectiveness of combining foundation model representations with efficient adaptation and hierarchical decoding for robust, scalable polyp segmentation.
Sanjana Jhansi Ganji, Panigrahi Srikanth, Kaushal Sambanna et al.· International Conference on...· 0 citations
Accurate colonoscopic polyp segmentation is essential for early colorectal cancer detection, yet remains challenging due to low contrast, specular highlights, blurred boundaries, and large variations in polyp appearance. We propose DFANet, a dual-branch feature aggregation framework that combines transformer-driven global semantic modeling MiT-B5 with CNN-based fine-grained representation learning EfficientNet-B5. A channel-spatial fusion module adaptively integrates multi-scale encoder outputs through element-wise aggregation, channel attention, and spatial attention, followed by a DSC Transformer that enhances contextual reasoning before decoding. The multi-stage decoder progressively restores resolution while refining structural details to produce accurate segmentation masks. Extensive experiments on five benchmark datasets demonstrate consistent performance gains, with DFANet achieving improvements of up to 3-7% in Dice and mIoU over state-of-the-art methods on Kvasir-SEG, CVC-ClinicDB, BKAI-IGH, ETIS-LaribPolypDB, and CVC-ColonDB.
Kokkanti Srinija, M. Harshini, Panigrahi Srikanth· International Conference on...· 0 citations
Background Colorectal cancer is a leading cause of cancer-related mortality, and reliable polyp segmentation during colonoscopy is critical for early intervention. Existing deep learning segmentors often produce blurred boundaries and are sensitive to appearance variation across endoscopy devices. Methods We propose BAR-UNet, a ResNet-34 encoder–decoder with a boundary-aware head and mask-guided appearance consistency learning (MACL). On the public Kvasir-SEG dataset we compare BAR-UNet with classical semantic segmentors (FCN-ResNet50, DeepLabV3-ResNet50), detection-based segmentors (YOLOv8-Seg, YOLOv11-Seg, Mask R-CNN), polyp-specific CNN baselines, and modern enhanced segmentors (DeepLabV3+, SegFormer-B2, MedSAM, SAM-Med2D) under a unified 70/10/20 split, multi-seed evaluation protocol. Results BAR-UNet achieves Dice 0.8798 ±0.0024 and IoU 0.7913 ± 0.0029 on Kvasir-SEG, outperforming all compared methods. Zero-shot evaluation on CVC-ClinicDB, CVC-ColonDB, and ETIS-Larib yields Dice/IoU of 0.831/0.748, 0.776/0.682, and 0.728/0.631, respectively. Ablation studies, boundary metrics (F1^bd, MAE, HD95), paired Wilcoxon tests, and MACL sensitivity analyses confirm that the boundary head and MACL are complementary. Discussion BAR-UNet improves boundary precision and appearance robustness with modest computational overhead on an NVIDIA RTX 4060 GPU. The method is a promising research prototype for computer-aided polyp analysis; prospective clinical validation is required before deployment.
The results suggest that running both models in parallel on the same input image could provide simultaneous classification and segmentation outputs, offering more comprehensive diagnostic information compared to single-task approaches.
Rahman Ardi Saputra, Suhendro Yusuf Irianto, Egi Safitri· Jurnal Nasional Pendidikan T...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
