Skin cancer detection with deep learning-based object detection models

n this study, four different deep learning–based object detection models—YOLOv8, YOLOv11, YOLOv12, and RT-DETR (Real-Time Detection Transformer)—were utilized to compare their performance in distinguishing between benign (non-malignant) and malignant skin lesions. The primary objective of the research is to conduct a comprehensive evaluation of these models in terms of accuracy, speed, generalization capability, and real-time performance. The ultimate goal is to contribute to the development of a faster, more reliable, and objective early diagnosis system that can be effectively implemented in clinical applications.