Precise segmentation of anatomical structures and pathological regions is crucial for diagnosis and treatment, especially in the field of medical imaging. In this work, we develop research to solve medical image segmentation by introducing semantic aligned matching (SAM) into the two popular transformer-based and real-time modules: Detection Transformers (DETRs), YOLOv8. Introduction: SAM, where the segmentation mask is aligned with image semantics to improve accuracy and robustness in medical image analysis it could be beneficial to guide the optimization indirectly through aligning our predictions more closely with semantic information encoded within an input data point. SAM-DETR and SAM-YOLO models for lung nodule segmentation were developed using LUNA16 dataset, giving significant improvements in performances. This integration is relevant because it shows promise of improved diagnostic and treatment possibilities through accurate segmentation in the field of medical imaging.