Comparative Analysis of Pre-Trained Deep Learning Models for Classifying Tropical Fungal Skin Infections

Suhardi Aras; Muhammad RIzwan Darwis; Muhammad Zhaky Arkan

doi:10.15575/join.v11i1.1758

Authors

Suhardi Aras Department of Informatics, Faculty of Engineering, Universitas Muhammadiyah Sorong, Indonesia
Muhammad RIzwan Darwis Department of Informatics, Faculty of Engineering, Universitas Muhammadiyah Sorong, Indonesia
Muhammad Zhaky Arkan Department of Informatics, Faculty of Engineering, Universitas Muhammadiyah Sorong, Indonesia

DOI:

https://doi.org/10.15575/join.v11i1.1758

Keywords:

Deep Learning, Fungal Skin Infections, Medical Image Classification, MobileNetV3, Tropical Skin Infections

Abstract

Tropical fungal skin infections, including Tinea corporis, Tinea versicolor, Tinea pedis, and Tinea nigra, are common health problems in tropical countries such as Indonesia. Although not life-threatening, these diseases can cause discomfort, reduce self-confidence, and interfere with daily activities. Conventional diagnostic methods still rely on subjective visual observation, which is often inaccurate—especially in regions with limited infrastructure and scarce access to specialists. Moreover, existing studies rarely provide a comparative evaluation of deep learning architectures for tropical fungal infections using small and diverse datasets. Therefore, this study aims to address these challenges by conducting a systematic comparative evaluation of three pre-trained models—MobileNetV3, EfficientNet-B2, and SE-ResNet101—to determine the most accurate and computationally efficient architecture for multi-class classification of tropical fungal skin diseases. In this study, a dataset of 660 clinical skin images sourced from the Kaggle repository was used, covering four tropical fungal infection classes. The dataset consisted of 165 images per class, which were divided into training, validation, and testing subsets. Experimental results demonstrated that MobileNetV3 achieved the best performance, with a validation accuracy of 95.08%, a test accuracy of 97.34%, and the shortest training time of 15 minutes, compared to EfficientNet-B2 (16 minutes) and SE-ResNet101 (22 minutes). The main contribution of this study is to provide a systematic comparative evaluation of deep learning models for the classification of tropical fungal skin infections, while recommending MobileNetV3 as the most suitable model for practical implementation of automated image-based diagnosis in primary healthcare services with limited resources.

References

[1] P. N. Srinivasu, J. G. Sivasai, M. F. Ijaz, A. K. Bhoi, W. Kim, and J. J. Kang, “Classification of skin disease using deep learning neural networks with mobilenet v2 and lstm,” Sensors, vol. 21, no. 8, Apr. 2021, doi: 10.3390/s21082852.

[2] D. Li et al., “Worldwide trends and future projections of fungal skin disease burden: a comprehensive analysis from the Global Burden of Diseases study 2021,” Front. Public Health, vol. 13, 2025, doi: 10.3389/fpubh.2025.1580221.

[3] World Health Organization (WHO), “Report of the first WHO global meeting on skin-related neglected tropical diseases,” World Health Organization (WHO), 2023. Accessed: Oct. 06, 2025. [Online]. Available: https://www.who.int/publications/i/item/9789240081201

[4] Ministry of Health and Republic of Indonesia (Kemenkes RI), “Laporan Nasional Riskesdas,” Ministry of Health, Republic of Indonesia (Kemenkes RI), Jakarta, 2019. Accessed: Jul. 06, 2025. [Online]. Available: https://www.litbang.kemkes.go.id/laporan-riskesdas-2019/

[5] S. Azzahra, E. Ervianti, and R. Setiabudi, “Clinical Patterns and Demographic Characteristics of Dermatophytosis in Surabaya,” Indonesian Journal of Tropical and Infectious Disease, vol. 12, no. 3, pp. 252–267, Dec. 2024, doi: 10.20473/ijtid.v12i3.66511.

[6] M. F. Petrucelli et al., “Epidemiology and diagnostic perspectives of dermatophytoses,” Journal of Fungi, vol. 6, no. 4, pp. 1–15, Dec. 2020, doi: 10.3390/jof6040310.

[7] V. Ravi, “Attention Cost-Sensitive Deep Learning-Based Approach for Skin Cancer Detection and Classification,” Cancers (Basel)., vol. 14, no. 23, Dec. 2022, doi: 10.3390/cancers14235872.

[8] M. K. Khalifaturohman, “Systematic Literature Review: Deep Learning Models in Arabic Script Classification,” Khazanah Journal of Religion and Technology, vol. 3, no. 1, pp. 18–22, Jul. 2025, doi: 10.15575/kjrt.v3i1.1619.

[9] M. Alruwaili and M. Mohamed, “AEfficientNetn Integrated Deep Learning Model with and ResNet for Accurate Multi-Class Skin Disease Classification,” Diagnostics, vol. 15, no. 5, Mar. 2025, doi: 10.3390/diagnostics15050551.

[10] A. Nur, A. Thohari, L. Triyono, I. Hestiningsih, B. Suyanto, and A. Yobioktobera, “Performance Evaluation of Pre-Trained Convolutional Neural Network Model for Skin Disease Classification,” JUITA: Jurnal Informatika, vol. 10, no. 1, pp. 9–18, 2022, Accessed: Oct. 06, 2025. [Online]. Available: https://jurnal.stmiktime.ac.id/index.php/juita/article/view/802

[11] T. Alyas, K. Alissa, A. S. Mohammad, S. Asif, T. Faiz, and G. Ahmed, “Innovative Fungal Disease Diagnosis System Using Convolutional Neural Network,” Computers, Materials and Continua, vol. 73, no. 3, pp. 4869–4883, 2022, doi: 10.32604/cmc.2022.031376.

[12] T. D. Nigat, T. M. Sitote, and B. M. Gedefaw, “Fungal Skin Disease Classification Using the Convolutional Neural Network,” J. Healthc. Eng., vol. 2023, 2023, doi: 10.1155/2023/6370416.

[13] T. Debebe and B. Molla, “Human Skin Fungal Diseases Classification Using Deep Learning Technique,” Haramaya Journal of Engineering and Technology (HJET), vol. 1, no. 2, pp. 16–37, 2022, [Online]. Available: https://journals.ddu.edu.et/index.php/HJET

[14] E. da S. Puls, M. V. Todescato, and J. L. Carbonera, “An evaluation of pre-trained models for feature extraction in image classification,” Oct. 2023, [Online]. Available: http://arxiv.org/abs/2310.02037

[15] P. Chanyachailert, C. Leeyaphan, and S. Bunyaratavej, “Cutaneous Fungal Infections Caused by Dermatophytes and Non-Dermatophytes: An Updated Comprehensive Review of Epidemiology, Clinical Presentations, and Diagnostic Testing,” Journal of Fungi, vol. 9, no. 6, Jun. 2023, doi: 10.3390/jof9060669.

[16] N. Łabędź, C. Navarrete-Dechent, H. Kubisiak-Rzepczyk, M. Bowszyc-Dmochowska, A. Pogorzelska-Antkowiak, and P. Pietkiewicz, “Pityriasis Versicolor—A Narrative Review on the Diagnosis and Management,” Oct. 01, 2023, Multidisciplinary Digital Publishing Institute (MDPI). doi: 10.3390/life13102097.

[17] Q. Yu et al., “Clinical Analysis of 57 Patients With Interdigital Infection in Shanghai, China: A Cross-Sectional Study,” Int. J. Dermatol. Venereol., vol. 8, no. 2, pp. 100–105, Jun. 2025, doi: 10.1097/JD9.0000000000000233.

[18] A. K. C. Leung, J. M. Lam, K. F. Leong, and K. L. Hon, “Tinea corporis: An updated review,” Jul. 20, 2020, Bioexcel Publishing LTD. doi: 10.7573/dic.2020-5-6.

[19] M. Trigka and E. Dritsas, “A Comprehensive Survey of Deep Learning Approaches in Image Processing,” Sensors (Basel), vol. 25, no. 2, Jan. 2025, doi: 10.3390/s25020531.

[20] D. Murcia-Gómez, I. Rojas-Valenzuela, and O. Valenzuela, “Impact of Image Preprocessing Methods and Deep Learning Models for Classifying Histopathological Breast Cancer Images,” Applied Sciences (Switzerland), vol. 12, no. 22, Nov. 2022, doi: 10.3390/app122211375.

[21] M. Tan and Q. V. Le, “EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks,” arXiv preprint arXiv:1905.11946, no. 6, Sep. 2020, doi: 10.48550/arXiv.1905.11946.

[22] A. Howard et al., “Searching for MobileNetV3,” in Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), IEEE, 2019. doi: 10.48550/arXiv.1905.02244.

[23] I. Manole, A. I. Butacu, R. N. Bejan, and G. S. Tiplica, “Enhancing Dermatological Diagnostics with EfficientNet: A Deep Learning Approach,” Bioengineering, vol. 11, no. 8, Aug. 2024, doi: 10.3390/bioengineering11080810.

[24] S. Thirumalaisamy et al., “Breast Cancer Classification Using Synthesized Deep Learning Model with Metaheuristic Optimization Algorithm,” Diagnostics, vol. 13, no. 18, Sep. 2023, doi: 10.3390/diagnostics13182925.

[25] B. Ashwini, M. Kaur, D. Singh, S. Roy, and M. Amoon, “Efficient Skip Connections-Based Residual Network (ESRNet) for Brain Tumor Classification,” Diagnostics, vol. 13, no. 20, Oct. 2023, doi: 10.3390/diagnostics13203234.

[26] M. Huang and Y. Xu, “Image classification of Chinese medicinal flowers based on convolutional neural network,” Mathematical Biosciences and Engineering, vol. 20, no. 8, pp. 14978–14994, 2023, doi: 10.3934/mbe.2023671.

[27] S. zhenlin, “Lightweight Face Anti-spoofing for Improved MobileNetV3,” Journal of Image Processing Theory and Applications, vol. 7, no. 1, 2024, doi: 10.23977/jipta.2024.070117.

[28] J. Mohan, A. Sivasubramanian, V. Sowmya, and R. Vinayakumar, “Enhancing Skin Disease Classification Leveraging Transformer-based Deep Learning Architectures and Explainable AI,” Comput. Biol. Med., Jul. 2024, doi: 10.1016/j.compbiomed.2025.110007.

[29] M. A. Khan, M. Sharif, T. Akram, R. Damaševičius, and R. Maskeliūnas, “Skin lesion segmentation and multiclass classification using deep learning features and improved moth flame optimization,” Diagnostics, vol. 11, no. 5, 2021, doi: 10.3390/diagnostics11050811.