PanTS: The Pancreatic Tumor Segmentation Dataset

Yang Yang (Nanjing University of Science and Technology) · Alan Yuille (JHU) · Tianyu Lin (Johns Hopkins University) · Qi Chen (Johns Hopkins University) · Kai Ding (Johns Hopkins University) · Zongwei Zhou (Johns Hopkins University) · Daguang Xu (NVIDIA) · Heng Li (SUN YAT-SEN UNIVERSITY) · Wenxuan Li (Johns Hopkins University) · Xinze Zhou (Johns Hopkins University) · Pedro R. A. S. Bassi (Johns Hopkins University, UniBo, IIT) · Xiaoxi Chen (University of Illinois at Urbana-Champaign) · Chen Ye (Peking University Third Hospital) · Zheren Zhu (UC Berkeley & UCSF) · Kang Wang (University of California, San Francisco) · Yucheng Tang (Karlsruher Institut für Technologie)
ai modelsanatomical structurescontrast phasein-plane spacinglarge-scale annotationslocalizationmetadatamulti-institutional datasetpancreatic ct analysispantsperformance evaluationsegmentationslice thicknesstumor detectionvoxel-wise annotations

PanTS is a large-scale, multi-institutional dataset curated to advance research in pancreatic CT analysis. It contains 36,390 CT scans from 145 medical centers, with expert-validated, voxel-wise annotations of over 993,000 anatomical structures, covering pancreatic tumors, pancreas head, body, and tail, and 24 surrounding anatomical structures such as vascular/skeletal structures and abdominal/thoracic organs. Each scan includes metadata such as patient age, sex, diagnosis, contrast phase, in-plane spacing, slice thickness, etc. AI models trained on PanTS achieve significantly better performance in pancreatic tumor detection, localization, and segmentation than those trained on existing public datasets. Our analysis indicates that these gains are directly attributable to the 16× larger-scale tumor annotations and indirectly supported by the 24 additional surrounding anatomical structures. As the largest and most comprehensive resource of its kind, PanTS offers a new benchmark for developing and evaluating AI models in pancreatic CT analysis.