🤖 AI Summary
Existing U-shaped medical image segmentation networks typically fuse multi-level features via element-wise addition or concatenation, which introduces redundancy and weakens cross-level complementarity, leading to inaccurate lesion localization and blurred boundaries. To address this, we propose the Multi-scale Intra–Multi-scale Subtraction Network (M2S-Net), featuring two key innovations: (1) a novel Multi-scale Intra-subtraction Unit (SU) and a pyramid-style cross-level multi-scale subtraction architecture that explicitly models complementary information through hierarchical differencing; and (2) a training-free LossNet that enables bottom-up, task-aware feature supervision. Evaluated across 11 cross-modality datasets—including colonoscopy, ultrasound, CT, and OCT—M2S-Net consistently outperforms state-of-the-art methods, achieving significant improvements in segmentation accuracy and boundary sharpness. This work establishes an efficient and interpretable paradigm for feature fusion in medical image segmentation.
📝 Abstract
Accurate medical image segmentation is critical for early medical diagnosis. Most existing methods are based on U-shape structure and use element-wise addition or concatenation to fuse different level features progressively in decoder. However, both the two operations easily generate plenty of redundant information, which will weaken the complementarity between different level features, resulting in inaccurate localization and blurred edges of lesions. To address this challenge, we propose a general multi-scale in multi-scale subtraction network (M$^{2}$SNet) to finish diverse segmentation from medical image. Specifically, we first design a basic subtraction unit (SU) to produce the difference features between adjacent levels in encoder. Next, we expand the single-scale SU to the intra-layer multi-scale SU, which can provide the decoder with both pixel-level and structure-level difference information. Then, we pyramidally equip the multi-scale SUs at different levels with varying receptive fields, thereby achieving the inter-layer multi-scale feature aggregation and obtaining rich multi-scale difference information. In addition, we build a training-free network ``LossNet'' to comprehensively supervise the task-aware features from bottom layer to top layer, which drives our multi-scale subtraction network to capture the detailed and structural cues simultaneously. Without bells and whistles, our method performs favorably against most state-of-the-art methods under different evaluation metrics on eleven datasets of four different medical image segmentation tasks of diverse image modalities, including color colonoscopy imaging, ultrasound imaging, computed tomography (CT), and optical coherence tomography (OCT). The source code can be available at url{https://github.com/Xiaoqi-Zhao-DLUT/MSNet}.
