EF-SwinNet: A Hybrid EfficientNet-Swin Transformer Model for Skin Cancer Classification
A hybrid EfficientNet + Swin Transformer architecture for skin lesion classification, reaching ~98% accuracy and a 96% F1 score on the HAM10000 dataset.
Publications
ADTNet: Attention-Guided U-Net with Dynamic CNN and Transformers for Skin Cancer Detection
An attention-driven U-Net variant combining dynamic CNNs and Transformers for skin cancer lesion segmentation, achieving a 92.4% Dice score and 87.4% IoU on ...