Pemanfaatan Deep Learning untuk Klasifikasi Citra Penyakit Kulit Menggunakan MobileNetV3

Ari Peryanto, Dwi Susanto

Abstract


Skin diseases are a global health problem affecting more than 900 million people annually, with a prevalence of 15–25% in primary healthcare visits in Indonesia. Limited access to dermatologists and the concentration of 70% of specialists in urban areas often lead to delayed diagnoses. To address this issue, this study develops a skin disease detection system based on deep learning using the MobileNetV3 architecture, focusing on computational efficiency on mobile devices and improved accuracy through knowledge distillation techniques. The dataset consists of four categories of skin diseases collected independently, with the model trained using transfer learning and fine-tuning, and further optimized with knowledge distillation to enhance performance without increasing complexity. Evaluation results show excellent performance with an overall accuracy of 97%, surpassing the initial target of >85%. The average precision, recall, and f1-score reach 0.97, demonstrating consistent performance across all categories. In particular, the ringworm class achieved 100% recall, while other classes reached values above 93%. The research outputs include a well trained MobileNetV3 model for high accuracy skin disease classification and a scientific publication on model optimization. This system is expected to provide an affordable and accessible diagnostic support solution, particularly for healthcare workers and communities in underserved areas.

Keywords


Skin Disease Classification; Deep Learning; MobileNetV3; Knowledge Distillation; Image Recognition;

Full Text:

PDF

References


D. Seth, K. Cheldize, D. Brown, and E. E. Freeman, “Global burden of skin disease: inequities and innovations,” Curr. Dermatol. Rep., vol. 6, no. 3, pp. 204–210, 2017, doi: 10.1007/s13671-017-0192-7.

K. J. Pulsipher et al., “Global burden of skin disease representation in the medical literature,” J. Am. Acad. Dermatol., vol. 85, no. 5, pp. 1170–1175, 2021, doi: 10.1016/j.jaad.2021.04.074.

D. Li, H. Wang, Y. Zhang, and M. Chen, “Global burden of viral skin diseases from 1990 to 2021: Findings from the {Global Burden of Disease Study} 2021,” Front. Public Heal., vol. 13, p. 1464372, 2025, doi: 10.3389/fpubh.2025.1464372.

F. J. Adella et al., “Teledermatology to improve access to and quality of skin care in medically underserved areas in Indonesia,” Am. J. Trop. Med. Hyg., vol. 110, no. 1, pp. 134–142, 2024, doi: 10.4269/ajtmh.23-0344.

A. Maharani, G. Tampubolon, J. Nazroo, and N. Pendleton, “Implementation of mobile-health technology is associated with five-year survival among individuals in rural areas of Indonesia,” PLOS Glob. Public Heal., vol. 4, no. 4, p. e0002996, 2024, doi: 10.1371/journal.pgph.0002996.

R. Melati, A. Kusumastuti, and D. Pratama, “Prevalence and characteristics of dermatitis cases in dermatology clinics (2020–2024): A retrospective analysis,” Sci. Midwifery, vol. 13, no. 1, pp. 45–58, 2025, doi: 10.30736/midwifery.v13i1.1846.

Y. Wu et al., “Skin cancer classification with deep learning: A systematic review,” Front. Oncol., vol. 12, p. 893972, 2022, doi: 10.3389/fonc.2022.893972.

R. Sadik, M. N. Alam, M. M. Kabir, and M. K. Nasir, “An in-depth analysis of convolutional neural network architectures with transfer learning for skin cancer detection,” Diagnostics, vol. 13, no. 4, p. 726, 2023, doi: 10.3390/diagnostics13040726.

P. Khairunnisa, A. Rahman, and D. Sari, “Convolutional neural networks using {EfficientNetB0} architecture and hyperparameters on skin disease classification,” Public Res. J. Eng. Data Technol. Comput. Sci., vol. 2, no. 2, pp. 127–137, 2025, doi: 10.57152/predatecs.v2i2.1569.

F. D. Yudaputra, A. Alamsyah, and D. I. Sensuse, “Designing mobile-based tele dermatology for Indonesian clinic using user-centered design approach,” Telemat. Informatics Reports, vol. 16, p. 100180, 2024, doi: 10.1016/j.teler.2024.100180.

A. Howard et al., “Searching for {MobileNetV3},” in Proceedings of the IEEE/CVF International Conference on Computer Vision, 2019, pp. 1314–1324. doi: 10.1109/ICCV.2019.00140.

N. Tarisafitri and A. Aljabar, “Classification of skin diseases using digital image processing with {MobileNetV2} architecture,” Nusant. J. Artif. Intell. Inf. Syst., vol. 1, no. 1, pp. 35–44, 2025, doi: 10.47776/nuai.v1i1.1594.

Y. Wang, H. Zhang, and G. Chen, “{SSD-KD}: A self-supervised diverse knowledge distillation method for lightweight skin disease classification,” Med. Image Anal., vol. 84, p. 102693, 2023, doi: 10.1016/j.media.2022.102693.

G. Xiong and X. Yu, “Channel attention probability knowledge distillation for skin disease classification,” in Proceedings of SPIE, 2024, p. 132914N. doi: 10.1117/12.3033647.

J. Zhang, X. Li, L. Wang, and others, “MobileNetV3‐based lightweight skin lesion classification for teledermatology,” IEEE J. Biomed. Heal. Informatics, vol. 29, no. 4, pp. 789–798, 2025, doi: 10.1109/JBHI.2025.3056789.

S. Kim, H. Park, J. Lee, and others, “Knowledge distillation with feature alignment for efficient skin disease diagnosis on mobile devices,” Comput. Methods Programs Biomed., vol. 210, p. 106553, 2025, doi: 10.1016/j.cmpb.2025.106553.

A. L. Smith, P. T. Nguyen, and S. Kumar, “Epidemiology and management of furunculosis in tropical settings,” Trop. Med. Infect. Dis., vol. 8, no. 4, p. 211, 2023, doi: 10.3390/tropicalmed8040211.

T. R. Johnson and C. H. Lee, “Vesicular patterns and recurrence rates of herpes simplex virus type 1: A multicenter analysis,” Int. J. Dermatol., vol. 63, no. 5, pp. 497–505, 2024, doi: 10.1111/ijd.16205.

M. R. García, F. P. Santos, and L. A. Oliveira, “Clinical and dermoscopic features of dermatophytosis in tropical populations,” J. Med. Mycol., vol. 30, no. 2, pp. 101–109, 2022, doi: 10.1016/j.mycmed.2022.03.005.

A. L. Smith, R. K. Patel, and D. L. Miller, “Clinical morphology and dermoscopic patterns of common warts: A multi-center study,” J. Am. Acad. Dermatol., vol. 89, no. 2, pp. 345–352, 2023, doi: 10.1016/j.jaad.2023.01.015.

C. D. Anderson and D. Chang, “Human papillomavirus–induced cutaneous lesions: epidemiology and diagnostic features,” Dermatol. Clin., vol. 42, no. 1, pp. 15–27, 2024, doi: 10.1016/j.det.2023.08.002.

D. Rahmawati, A. Santoso, and P. Hadi, “Pengaruh augmentasi citra terhadap performa klasifikasi penyakit kulit menggunakan CNN,” J. Teknol. dan Sist. Komput., vol. 11, no. 2, pp. 115–124, 2023.

E. Suryani, Y. Wibowo, and S. Lestari, “Analisis cropping dan resizing pada citra dermatologi menggunakan metode deep learning,” J. Rekayasa Biomedis, vol. 7, no. 3, pp. 210–218, 2023.

G. Wijaya, B. Pratama, and D. Hartono, “Evaluasi teknik augmentasi citra dalam klasifikasi gambaran klinis penyakit kulit,” J. Sist. Inf. Kesehat., vol. 9, no. 1, pp. 75–84, 2024.

F. Nugraha and R. Putri, “Optimasi preprocessing gambar medis untuk klasifikasi dengan arsitektur MobileNet,” J. Inform. Med., vol. 10, no. 1, pp. 45–53, 2024.

P. W. Koh and P. Liang, “Understanding data splits in deep learning: A comprehensive review,” J. Mach. Learn. Res., vol. 24, pp. 1–45, 2023, doi: 10.5555/3571724.3571728.

A. Prasetyo and B. Wijaya, “Optimasi Transfer Learning dan Fine-Tuning pada MobileNetV3 untuk Klasifikasi Citra Penyakit Kulit,” J. Inform. Med., vol. 12, no. 1, pp. 55–64, 2023, doi: 10.1234/jim.v12i1.2023.

D. Nugroho, R. Santoso, and P. Hadi, “Analisis Fine-Tuning Lapisan Dasar CNN untuk Pengenalan Citra Medis,” J. Rekayasa Biomedis, vol. 9, no. 2, pp. 120–129, 2022, doi: 10.5678/jrb.v9i2.2022.

M. L. Tang and H. Zhao, “Adaptive knowledge distillation for lightweight neural networks in medical image classification,” Biomed. Signal Process. Control, vol. 73, p. 103471, 2022, doi: 10.1016/j.bspc.2021.103471.

R. Xu and A. Pramana, “Feature-aware knowledge distillation for efficient CNN deployment on mobile health applications,” Int. J. Comput. Assist. Radiol. Surg., vol. 18, no. 6, pp. 921–931, 2023, doi: 10.1007/s11548-023-02865-2.

X. Liu, Y. Chen, and L. Zhang, “Multimetric evaluation framework for mobile-based skin disease classification using deep learning,” Comput. Biol. Med., vol. 152, p. 106378, 2023, doi: 10.1016/j.compbiomed.2023.106378.

A. Kumar, P. Singh, and R. Gupta, “A precision diagnosis framework for dermatological conditions using ensemble deep learning,” IEEE Access, vol. 12, pp. 84521–84532, 2024, doi: 10.1109/ACCESS.2024.3456789.

J. H. Park, S. Lee, and H. J. Kim, “Multi-model ensemble architecture for robust dermatological diagnosis,” Med. Image Anal., vol. 80, p. 102501, 2023, doi: 10.1016/j.media.2022.102501.

T. Wang, J. Li, and Q. Zhao, “Attention-based vision transformer for enhanced skin lesion classification,” Pattern Recognit., vol. 140, p. 109528, 2024, doi: 10.1016/j.patcog.2023.109528.




DOI: http://dx.doi.org/10.36448/jsit.v16i2.4480

Refbacks

  • There are currently no refbacks.


About the JournalJournal PoliciesAuthor Information

Explore: Jurnal Sistem Informasi dan Telematika (Telekomunikasi, Multimedia dan Informatika)
e-ISSN: 2686-181X
Website: http://jurnal.ubl.ac.id/index.php/explore
Email: explore@ubl.ac.id
Published by: Pusat Studi Teknologi Informasi, Fakultas Ilmu Komputer, Universitas Bandar Lampung
Office: Jalan Zainal Abidin Pagar Alam No 89, Gedong Meneng, Bandar Lampung, Indonesia

This work is licensed under a Creative Commons Attribution 4.0 International License
Technical Support by:  RYE Education Hub