TABNet: A Triplet Augmentation Self-Recovery Framework with Boundary-Aware Pseudo-Labels for Medical Image Segmentation

2025-07-03

AI Summary

To address insufficient feature learning and lack of boundary supervision caused by sparse scribble annotations in medical image segmentation, this paper proposes a weakly supervised learning framework. Methodologically, it introduces a triple-augmentation self-reconstruction module—incorporating intensity transformation, CutOut, and Jigsaw—to enable multi-granularity feature learning. A dual-branch prediction architecture with loss weighting is designed to generate high-quality pseudo-labels, while a boundary-aware loss function refines contour accuracy. Evaluated on the ACDC and MSCMR-seg benchmarks, the framework significantly outperforms existing weakly supervised approaches and achieves segmentation performance comparable to fully supervised models. This demonstrates its effectiveness in enhancing semantic understanding and boundary precision under sparse annotation constraints. The core contribution lies in the synergistic integration of augmented feature representation learning and boundary-guided pseudo-label optimization.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Semi-Supervised Learning
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

Background and objective: Medical image segmentation is a core task in various clinical applications. However, acquiring large-scale, fully annotated medical image datasets is both time-consuming and costly. Scribble annotations, as a form of sparse labeling, provide an efficient and cost-effective alternative for medical image segmentation. However, the sparsity of scribble annotations limits the feature learning of the target region and lacks sufficient boundary supervision, which poses significant challenges for training segmentation networks. Methods: We propose TAB Net, a novel weakly-supervised medical image segmentation framework, consisting of two key components: the triplet augmentation self-recovery (TAS) module and the boundary-aware pseudo-label supervision (BAP) module. The TAS module enhances feature learning through three complementary augmentation strategies: intensity transformation improves the model's sensitivity to texture and contrast variations, cutout forces the network to capture local anatomical structures by masking key regions, and jigsaw augmentation strengthens the modeling of global anatomical layout by disrupting spatial continuity. By guiding the network to recover complete masks from diverse augmented inputs, TAS promotes a deeper semantic understanding of medical images under sparse supervision. The BAP module enhances pseudo-supervision accuracy and boundary modeling by fusing dual-branch predictions into a loss-weighted pseudo-label and introducing a boundary-aware loss for fine-grained contour refinement. Results: Experimental evaluations on two public datasets, ACDC and MSCMR seg, demonstrate that TAB Net significantly outperforms state-of-the-art methods for scribble-based weakly supervised segmentation. Moreover, it achieves performance comparable to that of fully supervised methods.

Problem

• Addresses sparse scribble annotations in medical image segmentation
• Enhances feature learning with triplet augmentation strategies
• Improves boundary accuracy via pseudo-label supervision

Innovation

• Triplet augmentation enhances feature learning
• Boundary-aware loss refines segmentation contours
• Dual-branch predictions improve pseudo-label accuracy