Journal of Technology and Informatics
Comparative Evaluation of Deep Learning Architectures for Early Skin Cancer Detection Using Dermatoscopic Images
DOI:
https://doi.org/10.37802/Keywords:
Convolutional Neural Network, Dermoscopy, HAM10000, Skin Cancer, Transfer LearningAbstract
Early identification of skin cancer is essential for improving diagnostic reliability and patient outcomes. Although convolutional neural networks (CNNs) have demonstrated promising performance in dermoscopic image classification, comparative evaluations on imbalanced multiclass datasets remain challenging. This study compares three transfer learning-based CNN architectures—VGG16, ResNet50, and MobileNetV2—for seven-class skin lesion classification using the imbalanced HAM10000 dataset. All models were evaluated using the same dataset partitions, image resolution, and performance metrics, while architecture-specific augmentation and loss-function strategies were retained. In particular, VGG16 used MixUp, ResNet50 used CutMix, VGG16 and ResNet50 employed Focal Loss, and MobileNetV2 used label smoothing. Performance was assessed using accuracy, precision, recall, F1-score, confusion matrices, and ROC–AUC. ResNet50 achieved the best overall performance, with 85.71% accuracy and a weighted F1-score of 0.8502, ahead of VGG16 (82.52%, 0.8172) and MobileNetV2 (79.32%, 0.7739). ResNet50 also showed the highest recall for melanoma and dermatofibroma among the three tested configurations, although minority-class sensitivity remained limited. These results favor ResNet50 under the evaluated training configurations; however, differences in augmentation and loss functions prevent attributing performance gaps solely to network architecture.
Downloads
References
[1] H. Sung et al., "Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries," CA: A Cancer Journal for Clinicians, vol. 71, no. 3, pp. 209–249, 2021, doi: 10.3322/caac.21660.
[2] J. Yuan, X. Li, and S. Yu, "Global, Regional, and National Incidence Trend Analysis of Malignant Skin Melanoma Between 1990 and 2019, and Projections Until 2030," Cancer Control, vol. 31, pp. 1–11, 2024, doi: 10.1177/10732748241227340.
[3] R. L. Siegel, K. D. Miller, H. E. Fuchs, and A. Jemal, "Cancer Statistics, 2023," CA: A Cancer Journal for Clinicians, vol. 73, no. 1, pp. 17–48, 2023, doi: 10.3322/caac.21763.
[4] H. Bhatt, V. Shah, K. Shah, R. Shah, and M. Shah, "State-of-the-Art Machine Learning Techniques for Melanoma Skin Cancer Detection and Classification: A Comprehensive Review," Intelligent Systems with Applications, vol. 19, Art. no. 200290, 2023, doi: 10.1016/j.iswa.2023.200290.
[5] E. Ahishakiye, M. B. van Gijzen, J. Tumwiine, R. Wario, and J. Obungoloch, "A Survey on Deep Learning in Medical Image Reconstruction," Intelligent Medicine, vol. 1, no. 3, pp. 118–127, 2021, doi: 10.1016/j.imed.2021.08.004.
[6] A. Esteva, K. Chou, S. Yeung, N. Naik, J. Dean, and R. Socher, "Deep Learning-Enabled Medical Computer Vision," npj Digital Medicine, vol. 4, no. 5, 2021, doi: 10.1038/s41746-020-00376-2.
[7] A. Adegun and S. Viriri, "Deep Learning Techniques for Skin Lesion Analysis and Melanoma Cancer Detection: A Survey," Artificial Intelligence Review, vol. 53, pp. 6137–6176, 2020, doi: 10.1007/s10462-020-09865-4.
[8] K. He, X. Zhang, S. Ren, and J. Sun, "Deep Residual Learning for Image Recognition," in Proc. IEEE Conf. Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, USA, 2016, pp. 770–778, doi: 10.1109/CVPR.2016.90.
[9] M. Sandler, A. Howard, M. Zhu, A. Zhmoginov, and L.-C. Chen, "MobileNetV2: Inverted Residuals and Linear Bottlenecks," in Proc. IEEE/CVF Conf. Computer Vision and Pattern Recognition (CVPR), Salt Lake City, UT, USA, 2018, pp. 4510–4520, doi: 10.1109/CVPR.2018.00474.
[10] H. L. Gururaj, N. Manju, A. Nagarjun, V. M. N. Aradhya, and F. Flammini, "DeepSkin: A Deep Learning Approach for Skin Cancer Classification," IEEE Access, vol. 11, pp. 50205–50214, 2023, doi: 10.1109/ACCESS.2023.3274848.
[11] P. J., S. G., and R. Deepa, "Medical Image Classification for Skin Cancer Using Deep Learning and Ensemble Methods," in Proc. 2025 2nd International Conference on Artificial Intelligence and Knowledge Discovery in Concurrent Engineering (ICECONF), 2025, pp. 1–8, doi: 10.1109/ICECONF65644.2025.11379654.
[12] A. E. Minarno, A. Lusianti, Y. Azhar, and H. Wibowo, "Classification of Skin Cancer Images Using Convolutional Neural Network with ResNet50 Pre-trained Model," JOIV: International Journal on Informatics Visualization, vol. 8, no. 4, pp. 2013–2019, 2024.
[13] L. Li, X. Wang, W. Hu, Y. Du, B. Li, and N. N. Xiong, "Deep Learning in Skin Disease Image Recognition: A Review," IEEE Access, vol. 8, pp. 208264–208280, 2020, doi: 10.1109/ACCESS.2020.3037258.
[14] A. Mahbod, P. Tschandl, G. Langs, R. Ecker, and I. Ellinger, "The Effects of Skin Lesion Segmentation on the Performance of Dermatoscopic Image Classification," Computer Methods and Programs in Biomedicine, vol. 197, Art. no. 105725, 2020, doi: 10.1016/j.cmpb.2020.105725.
[15] C. Hernández-Pérez et al., "BCN20000: Dermoscopic Lesions in the Wild," Scientific Data, vol. 11, Art. no. 760, 2024, doi: 10.1038/s41597-024-03387-w.
[16] M. Musthafa, M. T. R., V. V., and S. Guluwadi, "Enhanced Skin Cancer Diagnosis Using Optimized CNN Architecture and Checkpoints for Automated Dermatological Lesion Classification," BMC Medical Imaging, vol. 24, 2024, doi: 10.1186/s12880-024-01356-8.
[17] M. Z. Khan, M. A. Khan, and N. N. Xiong, "Deep Neural Architectures for Medical Image Semantic Segmentation: Review," IEEE Access, vol. 9, pp. 83002–83024, 2021, doi: 10.1109/ACCESS.2021.3086530.
[18] H. S. R. Kumar, C. N. G. Babu, C. P. Vijay, K. Raju, K. L. S. Kumar, K. Prabhavathi, and K. Puttegowda, "Deep Learning-Based Automated Classification of Skin Lesions Using CNN and Computer Vision," SN Computer Science, vol. 6, 2025, doi: 10.1007/s42979-025-04375-6.
[19] K. Ali, Z. A. Shaikh, A. A. Khan, and A. A. Laghari, "Multiclass skin cancer classification using EfficientNets – a first step towards preventing skin cancer," Neuroscience Informatics, vol. 2, no. 1, Art. no. 100034, 2022, doi: 10.1016/j.neuri.2021.100034
[20] N. Gessert, M. Nielsen, M. Shaikh, R. Werner, and A. Schlaefer, "Skin Lesion Classification Using Ensembles of Multi-Resolution EfficientNets with Meta Data," MethodsX, vol. 7, Art. no. 100864, 2020, doi: 10.1016/j.mex.2020.100864.
[21] S. Serte and H. Demirel, "Wavelet-Based Deep Learning for Skin Lesion Classification," IET Image Processing, vol. 14, no. 4, pp. 720–726, 2020, doi: 10.1049/iet-ipr.2019.0553.
[22] P. Tschandl, C. Rosendahl, and H. Kittler, "The HAM10000 Dataset: A Large Collection of Multi-Source Dermatoscopic Images of Common Pigmented Skin Lesions," Scientific Data, vol. 5, Art. no. 180161, 2018, doi: 10.1038/sdata.2018.161.
[23] M. Xu, S. Yoon, A. Fuentes, and D. S. Park, "A Comprehensive Survey of Image Augmentation Techniques for Deep Learning," Pattern Recognition, vol. 137, Art. no. 109347, 2023, doi: 10.1016/j.patcog.2023.109347.
[24] M. Hammad, "Skin Cancer Diseases Classification Using Deep Convolutional Neural Network with Transfer Learning Model," Journal of Physics: Conference Series, vol. 2128, no. 1, Art. no. 012013, 2021, doi: 10.1088/1742-6596/2128/1/012013.
[25] A. A. Mamun, P. Ray, M. R. U. Nasib, A. Das, J. Uddin, and M. N. Absur, "Optimizing Deep Learning for Skin Cancer Classification: A Computationally Efficient CNN with Minimal Accuracy Trade-Off," in Proc. 2025 2nd International Conference on Next-Generation Computing, IoT and Machine Learning (NCIM), 2025, pp. 1–6, doi: 10.1109/NCIM65934.2025.11159853.
[26] A. M. Sarhan, H. A. Ali, S. Yasser, M. Gobara, A. Kandil, G. Sherif, and E. Moustafa, "Achieving High-Accuracy Skin Cancer Classification with Deep Learning Optimized by Ant Colony Algorithm," Journal of Electrical Systems and Information Technology, vol. 12, 2025, doi: 10.1186/s43067-025-00243-8.
