Skin cancer is a disease that is most prevalent and life threatening globally, where early diagnosis is very important in enhancing the survival chances of the patient. Nevertheless, the automation of dermoscopic image classification is still problematic because of the high similarity of images of various lesion types and the large inter-class variation. The Structural Similarity Index Measure (SSIM) was used to measure this difficulty on the 3297 images of the Dermoscopic skin lesion dataset, and the results showed that the similarity score (SSIM) was low among several samples in the same category, indicating the difficulty of the classification task. To overcome such problems, a hybrid deep-learning architecture is proposed to classify skin cancer, which uses various convolutional neural networks and fuzzy logic reasoning. Three pre-trained models, namely, DenseNet201, EfficientNetV2B2, and ResNet50, were used to extract features with the help of transfer learning. A number of hybrid models were developed by fusing the feature vectors of these networks. The experimental findings indicate that the tri-network hybrid model, which combines all three models, is better than each of the models and their combination. The proposed model had the overall accuracy of 0.91 and class-wise recall of 0.88 and 0.95, as well as the F1-score of 0.91 and 0.90 in benign and malignant classes, respectively. Moreover, the integration of a fuzzy logic inference system to narrow the prediction confidence enhanced the strength of the ultimate decision.
N. Uçar and M. F. Holick, “Illuminating the connection: cutaneous vitamin D3 synthesis and its role in skin cancer prevention,” Nutrients, vol. 17, no. 3, p. 386, 2025.
E. M. Cai, P. Pathmarajah, R. Daneshjou, J. M. Ko, and A. S. Chiou, “Evaluating the appropriateness of skin cancer prevention recommendations obtained from an online chat-based artificial intelligence model,” JAAD International, vol. 18, pp. 145-147, 2025.
B. Ozdemir and I. Pacal, “An innovative deep learning framework for skin cancer detection employing ConvNeXtV2 and focal self-attention mechanisms,” Results in Engineering, vol. 25, p. 103692, 2025.
D. Sridhar, “Deep Learning-Based Skin Cancer Detection: A CNN Approach for Early Diagnosis,” NCAI-DS 2025, p. 33, 2025.
S. Matiray and L. K. Singh, “Deep learning for early skin cancer detection: A comparative study on hybrid CNN models,” in Data, Information and Computing Science: Proceedings of the 2nd International Conference (CDICS 2024), Singapore, 6-8 December 2024, SAGE Publications, 2025, pp. 111-125.
S. K. Oruganti and U. Kumar, “Early Detection of Skin Diseases Using CNN-Based Detection Systems: A Comprehensive Review,” SGS-Engineering Science, vol. 1, no. 1, 2025.
S. Mavaddati, “Skin cancer classification based on a hybrid deep model and long short-term memory,” Biomedical Signal Processing and Control, vol. 100, p. 107109, 2025.
M. Bhargavi and S. Balakrishna, “Hybrid approach for multi-class skin cancer classification with DCNN feature and ensemble techniques,” Engineering Research Express, vol. 7, no. 3, p. 35260, 2025.
O. Yilmazer and O. K. Sahingoz, “Machine Learning Based Skin Cancer Detection Systems: A Comprehensive Survey,” in 2025 9th International Artificial Intelligence and Data Processing Symposium (IDAP), IEEE, 2025, pp. 1-8.
U. Zahra, A. Manan, M. Ali, A. Ahmad, and A. Hassan, “Early Detection and Multi-Class Classification of Skin Cancer Using Deep Learning and Transformer-Based Architectures,” Spectrum Engineering Science, vol. 4, no. 2, pp. 539-556, 2026.
V. Ranjan, K. Chaurasia, and J. Singh, “A comprehensive survey of skin cancer identification using deep learning,” Recent Advances in Electrical and Electronic Engineering, vol. 18, no. 9, pp. 1394-1412, 2025.
M. Haque, K. Khatun, G. Rohman, A. A. Rakib, and M. Billah, “A deep learning framework for multi-class skin cancer diagnosis: Addressing class imbalance and performance optimization through convolutional neural network architectures,” European Journal of Computer Science and Information Technology, vol. 13, no. 1, pp. 63-71, 2025.
I. Ashraf, M. Ullah, M. F. Nadeem, and M. N. Noor, “Coreset selection based on intra-class diversity,” arXiv preprint arXiv:2509.21380, 2025.
P. Natha, S. P. Tera, R. Chinthaginjala, S. O. Rab, C. V. Narasimhulu, and T. H. Kim, “Boosting skin cancer diagnosis accuracy with ensemble approach,” Scientific Reports, vol. 15, no. 1, p. 1290, 2025.
Hamsalekha R., “Fusion-Based Deep Learning Approach for Skin Cancer Detection Using DenseNet-121 and Transfer Learning,” Journal of Electrical Systems, vol. 20, no. 11s, pp. 3182-3188, 2024, doi: 10.52783/jes.8044.
A. A. Hussein, A. M. Montaser, and H. A. Elsayed, “Skin cancer image classification using hybrid quantum deep learning model with BiLSTM and MobileNetV2,” Quantum Machine Intelligence, vol. 7, no. 2, pp. 1-28, 2025, doi: 10.1007/s42484-025-00288-y.
H. Asgharnezhad, P. Tabarisaadi, A. Khosravi, R. Alizadehsani, and U. R. Acharya, “A Quad-Step Approach to Uncertainty-Aware Deep Learning for Skin Cancer Classification,” 2025, [Online]. Available: http://arxiv.org/abs/2506.10302.
W. Gouda, N. U. Sama, G. Al-Waakid, M. Humayun, and N. Z. Jhanjhi, “Detection of skin cancer based on skin lesion images using deep learning,” in Healthcare, MDPI, 2022, p. 1183.
M. Hosseinzadeh et al., “A model for skin cancer using combination of ensemble learning and deep learning,” PLOS ONE, vol. 19, no. 5, pp. 1-19, 2024, doi: 10.1371/journal.pone.0301275.
P. Natha and P. R. Rajeswari, “Skin Cancer Detection using Machine Learning Classification Models,” International Journal of Intelligent Systems and Applications in Engineering, vol. 12, no. 6s, pp. 139-145, 2024, doi: 10.20944/preprints202311.0248.v1.
M. Hasan, S. Das Barman, S. Islam, and A. W. Reza, “Skin cancer detection using convolutional neural network,” ACM International Conference Proceeding Series, pp. 254-258, 2019, doi: 10.1145/3330482.3330525.
R. Meghatria, D. Gaceb, F. Touazi, T. Boukert, and S. Ben Aidrene, “Enhancing Skin Cancer Detection Using Curriculum Learning in Ensemble Deep Learning Context,” vol. 10, 2026.
I. Bakurov, M. Buzzelli, R. Schettini, M. Castelli, and L. Vanneschi, “Structural similarity index (SSIM) revisited: A data-driven approach,” Expert Systems with Applications, vol. 189, p. 116087, 2022.
U. Gurav, R. Deshpande, and R. Mangulkar, “Soil-Type Detection: Comparative Analysis of VGG, ResNet-50, MobileNet, InceResNetV2, EfficientNet, DenseNet,” ICT Applications and Social Interfaces, Proceedings of ICTCS 2024, vol. 1323, p. 11, 2025.
Z. C. Oleiwi, E. N. AlShemmary, and S. Al-Augby, “Multi-label classification technique of chest X-rays image based cardiomegaly disease prediction,” in 2023 International Conference on Information Technology, Applied Mathematics and Statistics (ICITAMS), IEEE, 2023, pp. 134-139.
A. Hosna, E. Merry, J. Gyalmo, Z. Alom, Z. Aung, and M. A. Azim, “Transfer learning: a friendly introduction,” Journal of Big Data, vol. 9, no. 1, p. 102, 2022.
Z. Zhao, L. Alzubaidi, J. Zhang, Y. Duan, and Y. Gu, “A comparison review of transfer learning and self-supervised learning: Definitions, applications, advantages and limitations,” Expert Systems with Applications, vol. 242, p. 122807, 2024.
