Lightweight 2D Medical Image Segmentation Via a Decoder Using Linear Deformable Convolution and Multi-Scale...

High-precision medical image segmentation typically demands substantial computational resources, which poses a significant limitation to its practical deployment in resource-constrained regions. Consequently, the development of decoding mechanisms that are both computationally efficient and lightweight is imperative. However, the performance of medical image segmentation is frequently limited by the simplicity of decoder designs. Balancing the optimization of decoder architectures with the reduction of computational demands while maintaining high model accuracy remains a formidable challenge. In this study, we introduce LDMSD, a novel decoder that integrates linear deformable convolution and multi-scale self-attention. The multi-scale self-attention enhancement module within LDMSD leverages two distinct multi-scale self-attention mechanisms, thereby substantially improving the representational capacity of the feature maps. Furthermore, the decoder incorporates a linear deformable convolution attention-guided mechanism to augment the feature maps derived from skip connections. This mechanism effectively mitigates the inherent limitations of conventional convolution and enhances the model's ability to capture complex semantic relationships within the feature maps. Through this collaborative mechanism, LDMSD is able to capture target information from both global and multi-scale perspectives, accurately locate the target's boundaries and structures, while maintaining its lightweight nature. Experimental results demonstrate that LDMSD outperforms state-of-the-art decoders in terms of performance metrics, achieving a reduction in floating point operations (FLOPs) by 77.20% and in parameter count by 81.66% when compared to the CASCaded Attention DEcoder (CASCADE). The results on six publicly available datasets validate that the proposed method surpasses existing approaches in medical image segmentation tasks, both in terms of accuracy and computational efficiency.