Implicit Fuzzification via Bounded Noise Injection for Robust Medical Image Segmentation
For medical image segmentation, this method addresses boundary ambiguity without explicit fuzzy modeling, offering a simple improvement over standard U-Net.
NoiseUNet injects bounded noise into skip connections to regularize feature fusion, improving segmentation accuracy and boundary fidelity on medical images, including a new thyroid ultrasound dataset.
Image segmentation remains fundamentally limited by boundary ambiguity arising from sampling-induced information loss and inherent uncertainty in pixel-wise labeling. Although encoder-decoder architectures such as U-Net achieve strong performance, they often produce overconfident predictions that fail to capture transition-region ambiguity. To address this issue, we propose \textbf{NoiseUNet}, a simple yet effective framework that injects bounded perturbations into skip connections to regularize cross-scale feature fusion. This mechanism enforces robustness to local feature variations and promotes boundary-aware representations. Theoretically, the perturbation induces an implicit fuzzification effect, yielding soft, data-driven memberships without requiring explicit fuzzy modeling. We further introduce \textbf{ThyR}, a real-world thyroid ultrasound dataset with inherently ambiguous boundaries. Experiments demonstrate that NoiseUNet consistently improves both segmentation accuracy and boundary fidelity.