Purpose <p>Lung cancer remains one of the leading causes of cancer-related deaths, emphasizing the need for precise and effective detection methods. Deep learning-based object detection models, particularly the YOLO (You Only Look Once) series, have demonstrated promising results in medical imaging. This study aims to provide a comparative analysis of YOLOv7, YOLOv8, YOLOv9, and the latest YOLOv11 models for detecting lung nodules in CT scan images. The research evaluates their accuracy and confidence scores to determine their effectiveness in identifying lung nodules.</p> Methods <p>This study utilized deep learning-based YOLO models to detect lung nodules in CT scan images. The models were trained and tested using a dataset of CT scan images, and their performance was evaluated based on accuracy and confidence scores. The detection results of YOLOv7, YOLOv8, YOLOv9, and YOLOv11 were compared to assess their effectiveness in clinical applications.</p> Results <p>The comparative analysis revealed that both YOLOv11 and YOLOv8 achieved the highest accuracy of 93.33%, correctly classifying 14 out of 15 test images. However, YOLOv11 exhibited a higher confidence score, making it a more reliable choice for clinical applications. YOLOv9 achieved an accuracy of 80%, whereas YOLOv7 had the lowest accuracy at 66.66%.</p> Conclusions <p>The findings suggest that YOLOv11 outperforms other models in both accuracy and confidence, making it a strong candidate for automated lung cancer detection systems. This study highlights the significance of deep learning advancements in medical imaging and suggests further improvements in lung nodule detection methods to enhance early cancer diagnosis.</p>