Integrating YOLO for Advanced Content-Based Image Retrieval in Lung Cancer Imaging

This study proposes a Content-Based Image Medical Retrieval (CBIMR) system for lung cancer subtypes, integrating the TNM staging system. We employed YOLOv8, achieving top performance with a precision of 0.961, recall of 0.945, and mAP at 0.5 of 0.971. The model also provided high accuracy in retrieving similar cases across cancer groups and stages. This approach enhances diagnostic accuracy and interpretability, addressing challenges in current deep-learning algorithms for medical imaging.