|CT||DICOM, NIFTI, CSV||RSNA||Non-Commercial Use|
The RSNA 2022 Cervical Spine Fracture Detection Dataset is a large, meticulously organized dataset that facilitates the development of algorithms for identifying cervical spine fractures in CT scans. This kind of initiative is vital given the high incidence of spine fractures and the associated risk of neurologic deterioration and paralysis.
About Cervical Spine Fracture Dataset
The dataset includes CT studies from twelve sites across six continents, providing a diverse range of imaging data that can help develop generalizable algorithms. The annotations provided by spine radiology specialists from the ASNR and ASSR lend high credibility and accuracy to the dataset.
To create the ground truth dataset, the challenge planning task force collected imaging data sourced from twelve sites on six continents, including approximately 3,000 CT studies. With over 713,000 files and 343.51 GB of data, the dataset is robust enough to train complex machine learning models. The data is available in DICOM (
dcm), NIFTI (
nii), and CSV formats, providing flexibility in data manipulation and analysis. The dataset aims to improve the detection of cervical spine fractures, the most common site of spine fractures.