Dynamic iterative coarse-to-fine prompt learning based on SAM for precise esophageal cancer gross target volume segmentation.

Accurate segmentation of the gross tumor volume (GTV) from computed tomography (CT) images is a core step in the development of precise radiotherapy planning for esophageal cancer, which directly affects treatment efficacy and normal tissue protection. Recently, foundation models represented by the Segment Anything Model (SAM) have been increasingly applied in medical image segmentation, and several studies have extended them to tumor target segmentation with promising progress. However, existing SAM-based segmentation methods rely on manual prompts, leading to significant limitations in esophageal cancer. Manual localization of tumor prompt points is difficult, and inappropriate prompts easily cause segmentation errors, increasing the risk of damage to organs-at-risk (OAR) during radiotherapy. To address this, this study proposes an automatic segmentation method based on SAM. Inspired by the stepwise refinement of clinical target volume delineation, the core design of this method lies in a "coarse-to-fine" automatic prompt generation strategy. Specifically, coarse segmentation results generated by nnUNet are first converted into initial prompts to provide global anatomical priors for SAM. Furthermore, a dynamic iterative prompt update mechanism is introduced to construct a closed-loop system where optimized segmentation results drive prompt refinement, thereby gradually improving segmentation accuracy. Experimental results based on multi-center datasets show that this method achieves better segmentation performance than existing methods on both internal and external validation sets. It is highly consistent with the logic of clinical target volume delineation and can provide a reliable scheme for the formulation of precise radiotherapy plans for esophageal cancer.