Work

MedLiteNet: Lightweight Hybrid Medical Image Segmentation Model for the Medical Internet of Things (MIoT)

Year: 2025

Type: article

Abstract: Accurate skin-lesion segmentation remains a key technical challenge for computer-aided diagnosis of skin cancer. Convolutional neural networks, while effective, are constrained by limited receptive fields and thus struggle to model long-range dependencies. Vision Transformers capture global context, yet their quadratic complexity and large parameter budgets hinder use on the small-sample medical datasets common in dermatology. We introduce the MedLiteNet, a lightweight CNN-Transformer hybrid tailored for dermoscopic segmentation that achieves high precision through hierarchical feature extraction and multi-scale context aggregation. The encoder stacks depth-wise Mobile Inverted Bottleneck blocks to curb computation, inserts a bottleneck-level cross-scale token-mixing unit to exchange information between resolutions, and embeds a boundary-aware self-attention module to sharpen lesion contours. On the ISIC 2018 benchmark, a single MedLiteNet model attains a Dice score of$0.897 \pm 0.010$and an IoU of$0.821 \pm 0.015$with fewer than$3.3\,\mathrm{M}$parameters. A performance-weighted ensemble of three complementary variants raises accuracy to$0.904 \pm 0.012$Dice and$0.830 \pm 0.018$IoU while keeping the total parameter count below$10\,\mathrm{M}$-over$90\%$smaller than Vision-Transformer backbones. Qualitative results confirm superiority on irregular borders, low-contrast regions and multi-scale lesions, indicating MedLiteNet's suitability for real-time, resource-aware computer-aided dermatology.

Source: IEEE Journal of Selected Areas in Sensors

DOI: 10.1109/jsas.2025.3648685