Classification of Skin Disease from Skin Images Using Transfer Learning Technique

This article on skin disease classification using transfer learning. The proposed system combines image preprocessing, CNN-based feature extraction, ResNet50 transfer learning, Flask-based web deployment, Kafka streaming, and HBase storage. The design supports both academic experimentation and practical decision-support applications. The Phase II ResNet50 model achieved 95.8% accuracy, 95.2% precision, 94.9% recall, and 95.0% F1-score, outperforming VGG16 and MobileNetV2 baseline models. Accuracy graphs, loss curves, confusion matrices, system architecture, model summaries, and output screens demonstrate that the system learns meaningful features and can classify multiple skin disease images with high reliability. The system should be viewed as a preliminary screening aid rather than a replacement for professional medical diagnosis. With careful validation, privacy protection, and integration of explainable AI, it can become a useful component of telemedicine and intelligent healthcare systems.

[1] “Skin diseases image dataset,” Kaggle. Available: https://www.kaggle.com/datasets/ismailpromus/skin-diseases-image-dataset/data [2] “Studies on Different CNN Algorithms for Face Skin Disease Classification Based on Clinical Images,” IEEE Xplore, 2019. [3] E. Goceri, “Analysis of Deep Networks with Residual Blocks and Different Activation Functions: Classification of Skin Diseases,” 2019 Ninth International Conference on Image Processing Theory, Tools and Applications, 2019. [4] G. Singh, K. Guleria, and S. Sharma, “A Transfer Learning-based Pre-trained VGG16 Model for Skin Disease Classification,” MysuruCon, 2023. [5] N. Akmalia, P. Sihombing, and Suherman, “Skin Diseases Classification Using Local Binary Pattern and Convolutional Neural Network,” IEEE Xplore, 2019. [6] A. Esteva et al., “Dermatologist-level classification of skin cancer with deep neural networks,” Nature, 2017. [7] P. Tschandl, C. Rosendahl, and H. Kittler, “The HAM10000 dataset: A large collection of multi-source dermatoscopic images,” Scientific Data, 2018. [8] K. He, X. Zhang, S. Ren, and J. Sun, “Deep Residual Learning for Image Recognition,” Proc. IEEE CVPR, 2016. [9] J. Deng et al., “ImageNet: A Large-Scale Hierarchical Image Database,” Proc. IEEE CVPR, 2009. [10] F. Chollet, “Keras,” 2015. Available: https://keras.io