CTSpine1K: A Large-Scale Dataset for Spinal Vertebrae Segmentation in Computed Tomography
Yang Deng1,2,3, Ce Wang1,2, Yuan Hui1,2, Qian Li1,2, Jun Li1, Shiwei Luo4, Mengke Sun1, Quan Quan1, Shuxin Yang1, You Hao1,2, Pengbo Liu1, Honghu Xiao5, Chunpeng Zhao5, Xinbao Wu5, S. Kevin Zhou1,2,3
1: Key Lab of Intelligent Information Processing of Chinese Academy of Sciences, Institute of Computing Technology, Chinese Academy of Sciences, Beijing, 100190, China, 2: Suzhou Institute of Intelligent Computing Technology, Chinese Academy of Sciences, Suzhou, 215028, China, 3: School of Biomedical Engineering, Division of Life Sciences and Medicine, University of Science and Technology of China, Hefei, Anhui, and also with the Center for Medical Imaging, Robotics, Analytic Computing & Learning (MIRACLE), Suzhou Institute for Advanced Research, University of Science and Technology of China, Suzhou, Jiangsu, China, 4: Department of Radiology, Guangzhou First People's Hospital, School of Medicine, South China University of Technology, Guangzhou, 510006, China, 5: Beijing Jishuitan Hospital, Beijing, 100035, China
Publication date: 2025/12/31
https://doi.org/10.59275/j.melba.2025-gf84
Abstract
Spine-related diseases have high morbidity and cause a huge burden of social cost. Spine imaging is an essential tool for noninvasively visualizing and assessing spinal pathology. Segmenting vertebrae in computed tomography (CT) images has always been the base of quantitative medical image analysis for clinical diagnosis and surgery planning of spine diseases. Current publicly available annotated datasets on spinal vertebrae are small in size. Due to the lack of a large-scale annotated spine image dataset, the mainstream deep learning-based segmentation methods, which are data-driven, are heavily restricted. In this paper, we introduce a large-scale spine CT dataset called CTSpine1K, curated from multiple sources for vertebra segmentation, which contains 1,005 CT volumes with over 500,000 labeled vertebrae slices and 11,172 vertebrae belonging to different spinal conditions. Based on this dataset, we conducted several spinal vertebrae segmentation experiments to set the first benchmark. We believe that this large-scale dataset will facilitate further research in many spine-related image analysis tasks, including but not limited to vertebrae segmentation, labeling, 3D spine reconstruction from biplanar radiographs, and image superresolution and enhancement.
Our dataset are publically available at https://xnat.health-ri.nl/data/archive/projects/africai_miccai2024_ctspine1k and https://github.com/MIRACLE-Center/CTSpine1K.
Keywords
Spine Dataset · Vertebrae Segmentation · Computed Tomography (CT) · Medical Imaging
Bibtex
@article{melba:2025:037:deng,
title = "CTSpine1K: A Large-Scale Dataset for Spinal Vertebrae Segmentation in Computed Tomography",
author = "Deng, Yang and Wang, Ce and Hui, Yuan and Li, Qian and Li, Jun and Luo, Shiwei and Sun, Mengke and Quan, Quan and Yang, Shuxin and Hao, You and Liu, Pengbo and Xiao, Honghu and Zhao, Chunpeng and Wu, Xinbao and Zhou, S. Kevin",
journal = "Machine Learning for Biomedical Imaging",
volume = "3",
issue = "Special Issue on Open Data at MICCAI 2024–2025",
year = "2025",
pages = "824--832",
issn = "2766-905X",
doi = "https://doi.org/10.59275/j.melba.2025-gf84",
url = "https://melba-journal.org/2025:037"
}
RIS
TY - JOUR
AU - Deng, Yang
AU - Wang, Ce
AU - Hui, Yuan
AU - Li, Qian
AU - Li, Jun
AU - Luo, Shiwei
AU - Sun, Mengke
AU - Quan, Quan
AU - Yang, Shuxin
AU - Hao, You
AU - Liu, Pengbo
AU - Xiao, Honghu
AU - Zhao, Chunpeng
AU - Wu, Xinbao
AU - Zhou, S. Kevin
PY - 2025
TI - CTSpine1K: A Large-Scale Dataset for Spinal Vertebrae Segmentation in Computed Tomography
T2 - Machine Learning for Biomedical Imaging
VL - 3
IS - Special Issue on Open Data at MICCAI 2024–2025
SP - 824
EP - 832
SN - 2766-905X
DO - https://doi.org/10.59275/j.melba.2025-gf84
UR - https://melba-journal.org/2025:037
ER -