Large expert-curated datasets are essential for reliable scientific AI, but manual annotation remains slow and costly. In dentistry, detailed crown annotations are particularly valuable for computer-aided analysis of natural, restored, and prosthetic crowns. In this work, we present an expert-guided, algorithmically derived panoramic radiograph dataset of dental crown contours. The generation of dental crown annotations is characterized here as a hybrid human-AI workflow, involving the digital creation of precise segmentation masks for a dataset rather than the manual freehand delineation of each crown contour. Two radiologists annotated 769 panoramic radiographs in a customized PACS interface, and these expert CEJ/alveolar-ridge landmarks were then used as geometric guides to derive tooth-wise crown regions of interest stored as JSON annotations. We demonstrate the dataset's utility by training a YOLOv11x-Seg instance-segmentation model, chosen for its practical accuracy-speed trade-off and for producing object-level masks that facilitate downstream geometric analyses. The model achieved Mean Average Precision (mAP@.5:.95) of 70.5% for crown masks and 81.87% for the alveolar ridge. The dataset includes both alveolar ridge and CEJ annotations, supporting downstream CEJ-to-alveolar-ridge distance estimation. We also developed an end-to-end application that automates crown and alveolar ridge segmentation and derives these distances for expert review. In a perceptual study with three board-certified radiologists, our system's outputs received a favorable accuracy rating (perceived accuracy 70%) in 95.1% of the evaluated cases.