Hand MRI dataset

This website contains the world's first dataset of MRI scans of the human hand in multiple poses. Four subjects were scanned, each in 12 separate hand poses, for a total of 48 MRI scans in the dataset. The 12 poses are the same for all subjects. By scanning the same subject in 12 different poses, one can infer how the internal hand anatomy (bones, muscles, tendons, etc.) moves and deforms as the hand articulates.

The dataset is free (see license below). It was created at the University of Southern California by Bohan Wang (PhD candidate at USC), George Matcuk (Associate Professor of Radiology at USC) and Jernej Barbic (Professor of Computer Science at USC).

Scanning the hand in an arbitrary pose is not easy because the hand must be held perfectly still inside the MRI scanner for 10 minutes. We address this problem by manufacturing sturdy rubber-like molds that hold the hand in place in a specific pose during the scan. The method is described in our ACM SIGGRAPH 2019 paper.

License: The dataset is owned by the University of Southern California. It is licensed under the Attribution 4.0 International (CC BY 4.0) license. This license lets anyone distribute, remix, tweak, and build upon this work, even commercially, as long as they credit USC and the authors for the dataset. The specific legal text can be found on the Creative Commons website linked above. Please see "How to attribute / acknowledge" below for how to credit us.

Creative Commons License

How to attribute / acknowledge


Please use the following:

      Bohan Wang, George Matcuk, Jernej Barbic:
      Hand MRI dataset, University of Southern California, 2020.
We also provide a bibtex file.


If you would like to donate funding, please contact Jernej Barbic at jnb@usc.edu. The funds will be used to improve the modeling and understanding of the anatomy and function of the human hand.


  • [WMB19] Bohan Wang, George Matcuk, Jernej Barbič: Hand Modeling and Simulation Using Stabilized Magnetic Resonance Imaging, ACM Transactions on Graphics 38(4) (SIGGRAPH 2019), Los Angeles, CA, 2019
  • [WMB21] Bohan Wang, George Matcuk, Jernej Barbič: Modeling of Personalized Anatomy using Plastic Strains, ACM Transactions on Graphics (TOG), 40(2), 2021
  • [ZWHB22] Mianlun Zheng, Bohan Wang, Jingtao Huang, Jernej Barbič: Simulation of Hand Anatomy Using Medical Imaging, ACM Transactions on Graphics 41(6) (SIGGRAPH Asia 2022), Daegu, South Korea, 2022