MSDTCN-Net: A Multi-Scale Dual-Encoder Network for Skin Lesion Segmentation

TL;DR

This paper introduces MSDTCN-Net, a new network for skin lesion segmentation that improves accuracy by combining local and global features.

Contribution

The novel MSDTCN-Net integrates ConvNeXt, Deformable Transformer, and a multi-scale module for better skin lesion segmentation.

Findings

MSDTCN-Net outperforms existing methods on multiple skin lesion datasets.

The network achieves high IoU, Dice, and ACC metrics, showing strong generalization.

The HFT mechanism improves boundary reconstruction in lesion segmentation.

Abstract

Background/Objectives: Accurate segmentation of skin lesions is essential for early skin cancer detection. However, traditional CNNs are limited in modeling long-range dependencies, leading to poor performance on lesions with complex shapes. Methods: We propose MSDTCN-Net, a dual-encoder network that integrates ConvNeXt and Deformable Transformer to extract both local details and global semantic information. A Squeeze-and-Excitation (SE) mechanism is introduced to adaptively emphasize important channels. To address scale variation in lesions, we design a Multi-Scale Receptive Field (MSRF) module combining multi-branch and dilated convolutions. Furthermore, a Hierarchical Feature Transfer (HFT) mechanism is employed to guide high-level semantics progressively to shallow layers, enhancing boundary reconstruction in the decoder. Results: Extensive experiments on the ISIC 2016, ISIC 2017, ISIC 2018, and PH2 datasets show that MSDTCN-Net achieves competitive performance across metrics including IoU, Dice, and ACC, validating its effectiveness and generalization in skin lesion segmentation. Conclusions: MSDTCN-Net effectively combines local and global feature extraction, multi-scale adaptability, and semantic guidance to achieve high-accuracy skin lesion segmentation, demonstrating its potential in clinical diagnostic applications.

Figures