What question did this study set out to answer?

The goal is to enhance skin cancer detection accuracy by integrating various deep learning techniques for lesion classification.

April 24, 2026Open Access

From local textures to global attention: a hybrid feature fusion network for skin lesion classification

Key Points

The goal is to enhance skin cancer detection accuracy by integrating various deep learning techniques for lesion classification.
Implemented a unified feature-fusion framework utilizing preprocessing techniques like normalization and selective augmentation.
Utilized three deep learning architectures: VGG-16, ResNet-50, and Vision Transformer for feature extraction.
Employed HDFNet for classification on four public datasets: ISIC 2019, ISIC 2020, PAD-UFES, and DermQuest DERMIS.
Achieved a classification accuracy of 94.5% on the test datasets.
Obtained an area under the curve (AUC-ROC) score of 97%.
Outperformed existing models while providing interpretable predictions using Grad-CAM visual explanations.

Abstract

Accurate detection of skin cancer detection using RGB images remains a challenge due to multiple factors including variability in lesion appearance, difference in skin types, and the clinical interpretability of the models. To address these challenges, we present a unified feature-fusion framework that integrates deep learning methods for accurate classification of dermoscopic skin lesions. Our proposed method consists of preprocessing in which we performed normalization, class aware selective augmentation, followed by feature-level fusion from three customized deep learning architectures i.e.; VGG-16 with adaptive layer configuration, ResNet-50 with dermatological feature enhancement, and a Vision Transformer (ViT) with dynamic patching. These extracted features are then fused to form a comprehensive feature vector and classification is done using an HDFNet which is a two-layer Deep Neural Network (DNN). We have trained and tested the proposed model on four publicly available datasets including ISIC 2019, ISIC 2020, PAD-UFES, and DermQuest DERMIS. We achieved classification accuracy of 94.5% and an AUC-ROC of 97%. Our proposed method outperforms the existing state-of-the-art models and also provide interpretable predictions supported by Grad-CAM-based visual explanations.

Ask AI

Helpful

Bookmark

View Full Paper