The L2-Effnet for Human Skin Disease Multiclass Classification
Abstract
Skin diseases are common health problems that are often underestimated as most are mild and can be treated with over-the-counter medications, however, some types such as melanoma can be cancerous and deadly if not treated properly. Melanoma is caused by overexposure to ultraviolet light and has a 99% cure rate if diagnosed early, but decreases to 20% in advanced stages. In Indonesia, the uneven distribution of medical personnel and geographical challenges cause many cases of skin diseases not to be diagnosed well. This study develops a multi-class skin disease classification model using transfer learning by utilizing pre-trained models such as EfficientNetV2S to overcome data imbalance and improve accuracy. The dataset used is a combination of ISIC and Atlas Dermatology, which has been curated into 31 classes with a total of 3,399 samples, and data augmentation was performed to increase the sample size. The modified L2-EfficientNetV2S model showed a testing accuracy of 88.33%, higher than previous studies, demonstrating the potential of using deep learning in the early diagnosis of skin diseases to improve healthcare.
Downloads
References
Al-masni, M. A., Kim, D. H., & Kim, T. S. (2020). Multiple skin lesions diagnostics via integrated deep convolutional networks for segmentation and classification. Computer Methods and Programs in Biomedicine, 190, 105351. https://doi.org/10.1016/j.cmpb.2020.105351
Cahyanto, K. A., Yani, M., Rendi, Kustiyo, A., & Haryanto, T. (2023). The Effects of L2 Regularization in EfficientNet for Human Skin Disease Multi-Class Classification. Proceedings - 2023 6th International Conference on Computer and Informatics Engineering: AI Trust, Risk and Security Management (AI Trism), IC2IE 2023, 262–265. https://doi.org/10.1109/IC2IE60547.2023.10331108
Calderón, C., Sanchez, K., Castillo, S., & Arguello, H. (2021). BILSK: A bilinear convolutional neural network approach for skin lesion classification. Computer Methods and Programs in Biomedicine Update, 1, 100036. https://doi.org/10.1016/j.cmpbup.2021.100036
ISIC. (2019). ISIC challenge dataset. ISIC.
Isbahi, M. B., Zuana, M. M. M. ., & Mariana, E. R. . (2022). The Technology Strategy in Website Communication Media in Improving Business Activities. Majapahit Journal of Islamic Finance and Management, 1(2), 126–138. https://doi.org/10.31538/mjifm.v1i2.17
Jeong, H. K., Park, C., Henao, R., & Kheterpal, M. (2023). Deep Learning in Dermatology: A Systematic Review of Current Approaches, Outcomes, and Limitations. JID Innovations, 3(1), 100150. https://doi.org/10.1016/j.xjidi.2022.100150
Karthik, R., Vaichole, T. S., Kulkarni, S. K., Yadav, O., & Khan, F. (2022). Eff2Net: An efficient channel attention-based convolutional neural network for skin disease classification. Biomedical Signal Processing and Control, 73(October 2021), 103406. https://doi.org/10.1016/j.bspc.2021.103406
Khan, M. A., Muhammad, K., Sharif, M., & ... (2021). Multi-class skin lesion detection and classification via teledermatology. IEEE Journal of ….
Li, H., Pan, Y., Zhao, J., & Zhang, L. (2021). Skin disease diagnosis with deep learning: A review. Neurocomputing, 464, 364–393. https://doi.org/10.1016/j.neucom.2021.08.096
Purnama, I. K. E., Hernanda, A. K., Agung, A., Ratna, P., Nurtanio, I., & Hidayati, A. N. (2019). Disease Classification based on Dermoscopic Skin Images Using Convolutional Neural Network in Teledermatology System. 1–5.
Rafay, A., & Hussain, W. (2023). EfficientSkinDis : An EfficientNet-based classification model for a large manually curated dataset of 31 skin diseases. Biomedical Signal Processing and Control, 85(December 2022), 104869. https://doi.org/10.1016/j.bspc.2023.104869
Sadik, R., Majumder, A., Biswas, A. A., Ahammad, B., & Rahman, M. M. (2023). An in-depth analysis of Convolutional Neural Network architectures with transfer learning for skin disease diagnosis. Healthcare Analytics, 3(January), 100143. https://doi.org/10.1016/j.health.2023.100143
Setiawan, N. B., Natalia, F., Ferdinand, F. V., Sudirman, S., & Ko, C. S. (2021). Classification Of Skin Diseases And Disorders Using Convolutional Neural Network On A Mobile Application. ICIC Express Letters, 12(8), 715–721. https://doi.org/10.24507/icicelb.12.08.715
Shanthi, T., Sabeenian, R. S., & Anand, R. (2020). Automatic diagnosis of skin diseases using convolution neural network. Microprocessors and Microsystems, 76, 103074. https://doi.org/10.1016/j.micpro.2020.103074
Silva, S. F. da. (2025). Atlas Dermatology.
Soumare, H., Benkahla, A., & Gmati, N. (2021). Deep learning regularization techniques to genomics data. Array, 11(May), 100068. https://doi.org/10.1016/j.array.2021.100068
Tan, M., & Le, Q. V. (2021). EfficientNetV2: Smaller Models and Faster Training.
Zuana, M. M. M., Toha, M., & Isbahi, M. B. (2024). Exploration of Community Empowerment in a Village as the Entrance to a Lake in East Java. Malacca: Journal of Management and Business Development , 1(1), 47–55. https://doi.org/10.69965/malacca.v1i1.52
Copyright (c) 2026 Kurnia Adi Cahyanto, Kusworo Adi, Catur Edi Widodo
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgment of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgment of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work.
