<p>Lung cancer is the most common cancer worldwide and the primary contributor to global cancer mortality, so early and accurate diagnosis is particularly important. This paper introduces a new hybrid framework by fusing the YOLOv11 object detection model with a Large Language Model (LLM) for improved clinical interpretability. We trained YOLOv11 on labeled thoracic CT cases and tested for its ability to detect nodules and cancer, achieving a reported detection accuracy of 93.33% on a testing set of 150 images. YOLOv11 outputs precise (X, Y) coordinates with confidence scores. However, these results are not easily interpretable for clinicians. We address this by passing the detection outputs through an off-the-shelf explainable API to produce narrative-style radiology reports with recommended clinical follow-up. This process converts machine-detection data into human-readable, context-aware information that simulates the diagnostic process. The proposed YOLO-LLM framework connects AI outputs with medical interpretation and provides a powerful, explainable assistant for early lung cancer screening and clinical decision-making.</p>