🤖 AI Summary
To address low segmentation accuracy in 2D multi-organ medical image segmentation—caused by large inter-organ morphological variations and strong background interference—this paper proposes a lightweight, efficient spatially adaptive segmentation framework. Methodologically, we design a Spatially Adaptive Receptive Field Module (ARFM) that integrates deformable convolution DCNv3 with a Transformer-like architecture to model multi-scale local features; adopt a wide-and-shallow shared-parameter encoder-decoder structure to reduce parameter count; and introduce a Continuity-Aware Dynamic Weighting Loss that jointly optimizes boundary and region fidelity via an improved t-vMF Dice metric and cross-entropy. Evaluated on the Synapse 3D slice dataset, our method achieves significant improvements over state-of-the-art approaches in both segmentation accuracy and generalizability, while reducing model parameters by 32% and accelerating inference speed by 2.1×.
📝 Abstract
Multi-organ segmentation in medical image analysis is crucial for diagnosis and treatment planning. However, many factors complicate the task, including variability in different target categories and interference from complex backgrounds. In this paper, we utilize the knowledge of Deformable Convolution V3 (DCNv3) and multi-object segmentation to optimize our Spatially Adaptive Convolution Network (SACNet) in three aspects: feature extraction, model architecture, and loss constraint, simultaneously enhancing the perception of different segmentation targets. Firstly, we propose the Adaptive Receptive Field Module (ARFM), which combines DCNv3 with a series of customized block-level and architecture-level designs similar to transformers. This module can capture the unique features of different organs by adaptively adjusting the receptive field according to various targets. Secondly, we utilize ARFM as building blocks to construct the encoder-decoder of SACNet and partially share parameters between the encoder and decoder, making the network wider rather than deeper. This design achieves a shared lightweight decoder and a more parameter-efficient and effective framework. Lastly, we propose a novel continuity dynamic adjustment loss function, based on t-vMF dice loss and cross-entropy loss, to better balance easy and complex classes in segmentation. Experiments on 3D slice datasets from Synapse demonstrate that SACNet delivers superior segmentation performance in multi-organ segmentation tasks compared to several existing methods.