Annotated T2-weighted MR images of the Lower Spine

Annotated T2-weighted MR images of the Lower Spine Chengwen Chu, Daniel Belavy, Gabriele Armbrecht, Martin Bansmann, Dieter Felsenberg, and Guoyan Zheng Introduction The Institute for Surgical Technology and Biomechanics, University of Bern, Switzerland, Charité - University Medicine Berlin, Centre of Muscle and Bone Research, Free University & Humboldt-University Berlin, Germany, Centre for Physical Activity and Nutrition Research, School of Exercise and Nutrition Sciences, Deakin University Burwood Campus, Australia and Institut für Diagnostische und Interventionelle Radiologie, Krankenhaus Porz Am Rhein gGmbH, Köln, Germany, are making this dataset available as a resource in the development of algorithms and tools for spinal image analysis. Description The database consists of T2-weighted turbo spin echo MR spine images of 23 anonymized patients, each containing at least 7 vertebral bodies (VBs) of the lower spine (T11 – L5). For each vertebral body, reference manual segmentation is provided in the form of a binary mask. All images and binary masks are stored in the Neuroimaging Informatics Technology Initiative (NIFTI) file format, see details at http://nifti.nimh.nih.gov/. Image files are stored as "Img_xx.nii" while the associated annotation files are stored as "Img_xx_Labels.nii", where "xx" is the internal case number for the patient. Image annotations were prepared by Mr. Chengwen Chu (no professional training in radiology). Acknowledgements The acquisition of original images was supported by the Grant 14431/02/NL/SH2 from the European Space Agency, grant 50WB0720 from the German Aerospace Center (DLR) and the Charité Universitätsmedizin Berlin. Preparation of this data collection was made possible thanks to the funding from the Swiss National Science Foundation (SNSF) through project: 205321 157207/1. Reference C. Chu, D. Belavy, W. Yu, G. Armbrecht, M. Bansmann, D. Felsenberg, and G. Zheng, “Fully Automatic Localization and Segmentation of 3D Vertebral Bodies from CT/MR Images via A Learning-based Method”, PLoS One. 2015 Nov 23;10(11):e0143327. doi: 10.1371/journal.pone.0143327. eCollection 2015.