Skin Cancer Classification Using VGG-16

Melanoma is a highly fatal form of skin cancer, where early and accurate diagnosis plays a vital role in reducing mortality. Due to the striking similarities among different types of skin lesions, manual diagnosis remains challenging. Dermatologists rely on early-stage classification of skin lesions to administer timely treatment and save lives. This paper presents an effective deep learning-based classification model utilizing the VGG16 architecture through transfer learning. The proposed model successfully differentiates between benign and malignant skin lesions using a dataset comprising 1,800 benign and 1,498 malignant skin images collected from online sources. The model achieves a training accuracy of 99.62% and a validation accuracy of 84.97%, highlighting its potential for reliable clinical support.