Enhancing Social Impact Through AI: CNN-Based Early Melanoma Detection for Improving Skin Cancer Diagnostics and Healthcare Access

Skin cancer, particularly melanoma, remains a critical global health challenge where early detection significantly improves patient survival and reduces healthcare costs. In response to this need, this study presents a deep learning-based approach to transforming skin cancer diagnosis through accessible, accurate automated detection. A convolutional neural network (CNN) architecture with YOLO11 was developed and evaluated to improve melanoma classification in dermatological images. To further enhance diagnostic reliability, an image preprocessing stage was implemented to remove artifacts such as hair, thereby improving overall image quality and classification performance. The proposed model achieved 90% accuracy, with strong sensitivity and specificity, demonstrating its effectiveness in distinguishing benign from malignant lesions. Beyond technical performance, this work highlights the potential of CNN to expand access to early melanoma detection, particularly in resource-limited settings where specialist care may be scarce. By enabling faster, more cost-effective, and more reliable screening, this approach helps reduce diagnostic delays, optimize healthcare resources, and ultimately improve patient outcomes and quality of life. These results support the continued development of AI-driven tools as scalable solutions for equitable and impactful healthcare delivery.