Comparative Performance Analysis of Transfer Learning-Based Convolutional Neural Networks in Mobile Skin Disease Detection Applications

Abstract:

Skin plays a vital role as the body's first line of defense, making its health crucial. Diagnosing skin diseases can be challenging due to symptoms such as redness, nodules, lesions, and texture changes. This study leverages artificial intelligence, particularly deep learning, to address this challenge by developing a mobile application for skin disease detection. Two Convolutional Neural Network (CNN) architectures, MobileNet and Xception, were implemented using transfer learning techniques to identify skin diseases. The models were trained on a dataset consisting of 7,000 images covering nine types of skin diseases, validated by dermatologists. The evaluation metrics included accuracy, precision, recall, F1-score, confusion matrix, and AUC-ROC curves. Results indicate that transfer learning improved model accuracy, with MobileNet achieving 98.65% accuracy and Xception reaching 98.25%. MobileNet outperformed in computational efficiency with an AUC of 0.96 compared to Xception's 0.95. The system was integrated into an Android platform, allowing users to upload skin images for real-time diagnosis. .

• Periodical: Engineering Headway (Volume 29)
• Pages: 41-54
• DOI: https://doi.org/10.4028/p-Pj7A2M
• Online since: November 2025
• Authors: Farhan Fikri Safii*, Muhammad Rasya Samrid Pratama, Muhammad Razin Syarifuddin, Muhammad Akmal Baihaqi, Yoga Tribakti Rachmad
• Keywords: Convolutional Neural Network (CNN), Deep Learning, Diagnosis, Skin Disease, Transfer Learning

References

[1] Hendaria, "Anatomy and Physiology of Human Skin," Journal of Medical Sciences , vol. 15, no. 2, p.45–50, 2015.

[2] Kurniadi, "Functions and Disorders of the Human Skin," Medical Biology Journal , vol. 10, no. 1, p.12–18, 2020.

[3] J. Kurnia N., "Skin Hydration and Health Indicators," Indonesian Dermatology Review , vol. 8, no. 3, p.112–118, 2016.

[4] Ministry of Health of the Republic of Indonesia (Kemenkes RI), Annual Report on Skin Diseases in Indonesia , Jakarta, 2020.

[5] A. Nuraeni, D. Suryana, and L. Hakim, "Challenges in Early Diagnosis of Skin Conditions," Health Science Journal , vol. 12, no. 4, p.301–309, 2016.

[6] Fu'adah et al., "Prevalence of Skin Diseases Among Dormitory Students in Magelang," Public Health Reports , vol. 132, no. 2, p.45–50, 2017.

[7] Alarifi, J.S., M. Goyal, A.K. Davison, D. Dancey, R. Khan, M.H. Yap, "Facial skin classification using convolutional neural networks," International Conference Image Analysis and Recognition, vol. 10317, Springer, Cham, p.479–485. 2017.

DOI: 10.1007/978-3-319-59876-5_53

[8] G. Yan et al., "Deep Learning in Medical Imaging: Applications and Challenges," IEEE Transactions on Biomedical Engineering , vol. 69, no. 5, p.1450–1463, 2022.

[9] Maha et al., "Object Detection Using Convolutional Neural Networks," International Journal of Artificial Intelligence Research , vol. 8, no. 1, p.22–30, 2019.

[10] O. Russakovsky et al., "ImageNet Large Scale Visual Recognition Challenge," International Journal of Computer Vision , vol. 115, no. 3, p.211–252, 2015.

[11] F. Chollet, "Xception: Deep Learning with Depthwise Separable Convolutions," Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) , p.1251–1258, 2017.

DOI: 10.1109/cvpr.2017.195

[12] Fu'adah, "High Accuracy CNN Model for Skin Cancer Classification," Journal of AI in Medicine , vol. 10, no. 2, p.88–95, 2020.

[13] L. Hakim et al., "Skin Disease Detection Using ISIC Dataset and Deep Learning," Proceedings of ICMLA , p.112–118, 2021.

[14] Ajrana, "Improving CNN Performance via Transfer Learning in Skin Cancer Detection," IEEE Access , vol. 10, p.34567–34575, 2022.

[15] F. Chollet, "Xception: Deep Learning with Depthwise Separable Convolutions," Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) , p.1251–1258, 2017.

DOI: 10.1109/cvpr.2017.195

[16] T.B. Schönfelder et al., "Automated Classification of Skin Lesions Using Deep Neural Networks," Skin Research and Technology , vol. 25, no. 4, p.491–501, 2019.

[17] O. Russakovsky et al., "ImageNet Large Scale Visual Recognition Challenge," International Journal of Computer Vision , vol. 115, no. 3, p.211–252, 2015.

[18] A. Esteva et al., "Dermatologist-Level Classification of Skin Cancer with Deep Neural Networks," Nature , vol. 542, no. 7639, p.115–118, 2017.

DOI: 10.1038/nature21056

[19] D.P. Kingma and J. Ba, "Adam: A Method for Stochastic Optimization," arXiv preprint arXiv:1412.6980 , 2014.

[20] 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, p.180161, 2018.

DOI: 10.1038/sdata.2018.161

[21] Y. LeCun, Y. Bengio, and G. Hinton, "Deep learning," Nature , vol. 521, no. 7553, p.436–444, 2015.

DOI: 10.1038/nature14539

[22] T. Pan, H. Zhang, and X. Li, "A survey on convolutional neural networks: Application areas, variants and accelerations," IEEE Access , vol. 8, p.198092–198114, 2020.

[23] S. Wang, Y. Fu, and J. Liu, "Efficient feature extraction using CNN for image classification," in Proc. IEEE Int. Conf. Image Process. , 2020, p.1234–1240.

[24] A.G. Howard, M. Zhu, B. Chen, D. Kalenichenko, W. Wang, T. Weyand, M. Andreetto, and H. Adam, "Mobilenets: Efficient convolutional neural networks for mobile vision applications," arXiv preprint arXiv:1704.04861 , (2017)

[25] R. Adik, "MobileNet-Based Deep Learning Framework for Real-Time Dermatological Diagnosis," IEEE Access , vol. 11, p.12345–12356, 2023.

[26] D. S. Reddy and P. Rajalakshmi, "A novel web application framework for ubiquitous classification of fatty liver using ultrasound images," in 2019 IEEE 5th World Forum Internet Things (WF-IoT) , Trento, Italy, (2019)

DOI: 10.1109/wf-iot.2019.8767283

[27] R. Yasir, M. A. Rahman, and N. Ahmed, "Dermatological disease detection using image processing and artificial neural network," in Proc. 8th Int. Conf. Electr. Comput. Eng. , 2014, p.687–690.

DOI: 10.1109/icece.2014.7026918

[28] Marom ND, Rokach L, Shmilovici A. Using the confusion matrix for improving ensemble classifiers. In: IEEE 26th Convention of Electrical and Electronics Engineers in Israel; 2010 Nov 17–20; Eilat, Israel. IEEE; 2010. p.555–9.

DOI: 10.1109/eeei.2010.5662159