This paper proposes a hybrid method for skin lesion classification combining deep learning features with conventional descriptors such as HOG, Gabor, SIFT, and LBP. Feature extraction was performed by extracting features of interest within the tumor area using suggested fusion methods. We tested and compared features obtained from different deep learning models coupled to HOG-based features. Dimensionality reduction and performance improvement were achieved by Principal Component Analysis, after which SVM was used for classification. The compared methods were tested on the reference database skin cancer-malignant-vs-benign. The results show a significant improvement in terms of accuracy due to complementarity between the conventional and deep learning-based methods. Specifically, the addition of HOG descriptors led to an accuracy increase of 5% for EfficientNetB0, 7% for ResNet50, 5% for ResNet101, 1% for NASNetMobile, 1% for DenseNet201, and 1% for MobileNetV2. These findings confirm that feature fusion significantly enhances performance compared to the individual application of each method.

Skin lesion classification is a significant process in the diagnosis and early detection of melanoma and other conditions. In this study, a combined technique based on LSTM and GRU models was applied to extract spatial and sequential features from skin cancer images and to improve their classification into benign or malignant classes. In order to improve classification performance, various architectures are explored, including BiLSTM, GRU, and GRU combined to BiLSTM. These models process image data as temporal sequences, so that it can learn complex dependencies between pixel patterns. The conducted experiments display that the GRU-based method reached an accuracy of 81.21%, compared to 81% for the BiLSTM model. The blend of both architectures outperformed each of the two individual models, registering a higher accuracy of 85%, thus showcasing the potential of the hybrid approach. These results indicate the strengths of LSTM-GRU combined model in untying temporal structures from sequential features and improving classification accuracy when fused for deep feature extraction. The application of Batch Normalization and Dropout also assists in creating a stable, accurate, and robust classification system.

Skin cancer is one of the most common diseases worldwide and, if left untreated, it can be life threatening. In this work, we propose a dual-branch deep learning framework that integrates a CNN–BiLSTM module for local spatial–sequential feature modeling with multiple transformer models (ViT, DeiT, SwinV2, and BEiT) for global representation learning. Each input image is processed in parallel by two branches. The first branch consists of a lightweight convolutional network, followed by row-wise and column-wise Bidirectional LSTM layers to capture spatial and sequential dependencies. The second branch employs a high-resolution transformer to extract features from the image. The features extracted from both branches of the hybrid CNN–BiLSTM–Transformer model are fused to construct a unified classifier for accurate skin lesion recognition. To further improve robustness, four CNN–BiLSTM–Transformer hybrid models are combined using a stacking strategy based on logistic regression that aggregates their prediction probabilities. Experimental results on the binary Kaggle and multi-class HAM10000 datasets demonstrate that stacking achieves better performance than the individual hybrid CNN–BiLSTM–Transformer models, reaching accuracies of 91.21% on the binary Kaggle dataset and 86.08 ± 0.47% on the HAM10000 dataset, respectively. These findings confirm the efficiency of using complementary local and global features for skin lesion classification.