Abstract: Melanoma is one of the deadliest forms of skin cancer, and early diagnosis is critical for improving patient survival rates. This paper presents a deep learning-based melanoma detection system that classifies dermoscopic skin images into benign and malignant categories. The proposed system employs a Convolutional Neural Network (CNN) with EfficientNet architecture for accurate feature extraction and classification. A Flask-based web application is developed to enable users to upload images and receive real-time predictions. The experimental results demonstrate that the proposed approach achieves reliable accuracy and can assist dermatologists in clinical decision-making The proposed system employs a Convolutional Neural Network (CNN) using EfficientNet architecture to classify dermoscopic skin lesion images into benign and malignant categories. Image preprocessing techniques including resizing, normalization, and data augmentation are applied to enhance model robustness and reduce overfitting. The trained model is integrated into a web-based application using the Flask framework, enabling users to upload skin lesion images and receive real-time prediction results.

Keywords: Melanoma Detection, Deep Learning, CNN, EfficientNet, Medical Image Analysis

Downloads: | DOI: 10.17148/IJIREEICE.2025.131211