Authors : Vanitha Muthu P.; Bhuvaneswari C.; Meenushree R.; Rajasri R.; Vinisha E.

Volume/Issue : Volume 11 - 2026, Issue 4 - April

Google Scholar : https://tinyurl.com/bdh7mp24

Scribd : https://tinyurl.com/zx99xau3

DOI : https://doi.org/10.38124/ijisrt/26apr1153

Note : A published paper may take 4-5 working days from the publication date to appear in PlumX Metrics, Semantic Scholar, and ResearchGate.

Abstract : Accurate and timely diagnosis of skin diseases is essential for enabling early treatment, reducing healthcare costs, and improving patient outcomes. Conventional diagnostic approaches rely on manual visual inspection by dermatologists, which can be subjective, time-consuming, and often inaccessible in remote or resource-limited settings. This paper presents DermAnalyze AI, a multi-stage framework that integrates image processing with lightweight deep learning for effective skin disease classification and severity estimation. The proposed system incorporates a novel gatekeeper mechanism that initially verifies whether the input image contains human skin using colour-space filtering, thereby rejecting irrelevant or low-quality inputs and ensuring diagnostic integrity. For validated inputs, a MobileNetV2-based architecture is employed for efficient feature extraction and classification of various skin conditions. In addition, a dedicated severity analysis module computes a quantitative severity index, categorizing cases into mild, moderate, and severe levels. This severity-aware design extends beyond conventional classification by providing clinically interpretable insights and supporting treatment recommendations. Experimental evaluation on the HAM10000 dataset demonstrates that the proposed model achieves an accuracy of 93.2% and a weighted F1-score of 0.92, indicating robust classification performance across multiple disease categories. Furthermore, the lightweight architecture ensures reduced computational complexity and faster inference, making it suitable for deployment in mobile and resource-constrained environments. Overall, the proposed framework offers a scalable, reliable, and accessible solution for real-world dermatological screening.

Keywords : Skin Disease Classification, Severity Estimation, Deep Learning, MobileNetV2, Medical Image Analysis, Computer Vision, HAM10000 Dataset, Clinical Decision Support, Lightweight Neural Networks, Mobile Healthcare

References :

1. D. Makolo, et al., “An optimized deep learning-based system for accurate detection and classification of skin diseases,” International Journal of Research and Innovation in Applied Science (IJRIAS), vol. 10, no. 2, pp. 145–160, 2025.
2. P. N. Srinivasu, et al., “Classification of skin disease using deep learning neural networks with MobileNetV2 and LSTM,” Sensors, vol. 21, no. 8, pp. 2852–2870, 2021.
3. L. F. Li and H. Wang, “Deep learning in skin disease image recognition: A review,” IEEE Access, vol. 8, pp. 165287–165299, 2020.
4. S. Ahmed, et al., “Deep-learning-based super-resolution and classification framework for skin disease detection,” Optical and Quantum Electronics, vol. 54, no. 11, pp. 750–765, 2022.
5. S. Saiwaeo and N. Phanthumchinda, “Human skin type classification using image processing and deep learning approaches,” Heliyon, vol. 9, no. 4, pp. 152–165, 2023.
6. P. Tschandl, C. Rosendahl, and H. Kittler, “The HAM10000 dataset: A large collection of multi-source dermatoscopic images,” Scientific Data, vol. 5, no. 1, pp. 1–9, 2018.
7. A. Esteva, B. Kuprel, R. A. Novoa, et al., “Dermatologist-level classification of skin cancer with deep neural networks,” Nature, vol. 542, no. 7639, pp. 115–118, 2017.
8. N. Codella, D. Gutman, M. Celebi, et al., “Skin lesion analysis toward melanoma detection: A challenge at the International Symposium on Biomedical Imaging,” IEEE Journal of Biomedical and Health Informatics, vol. 23, no. 2, pp. 501–512, 2019.
9. M. Tschandl, C. Rosendahl, and H. Kittler, “The HAM10000 dataset: A large collection of multi-source dermatoscopic images,” Scientific Data, vol. 5, no. 1, pp. 1–9, 2018.
10. 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), 2016, pp. 770–778.
11. M. Sandler, A. Howard, M. Zhu, A. Zhmoginov, and L. Chen, “MobileNetV2: Inverted residuals and linear bottlenecks,” in Proc. IEEE Conf. Computer Vision and Pattern Recognition (CVPR), 2018, pp. 4510–4520.
12. F. Yu, V. Koltun, and T. Funkhouser, “Dilated residual networks,” in Proc. IEEE Conf. Computer Vision and Pattern Recognition (CVPR), 2017, pp. 472–480.
13. G. Litjens, T. Kooi, B. E. Bejnordi, et al., “A survey on deep learning in medical image analysis,” Medical Image Analysis, vol. 42, pp. 60–88, 2017.
14. H. Nasr-Esfahani, S. Samavi, N. Karimi, et al., “Melanoma detection by analysis of clinical images using convolutional neural network,” in Proc. IEEE Int. Conf. Image Processing (ICIP), 2016, pp. 137–141.
15. P. Tschandl, C. Rosendahl, and H. Kittler, “Expert-level diagnosis of nonpigmented skin cancer by combined convolutional neural networks,” JAMA Dermatology, vol. 155, no. 1, pp. 58–65, 2019.
16. J. Deng, W. Dong, R. Socher, et al., “ImageNet: A large-scale hierarchical image database,” in Proc. IEEE Conf. Computer Vision and Pattern Recognition (CVPR), 2009, pp. 248–255.