This is the demo datasets that can be used with the open-sourced code for the Motion-Robust Dental MRI: https://github.com/ZihanNing/dental_MRI_motion_correction Raw data: meas_MID03419_FID159770_PDwSPACE_DISORDER_ETL30_move.dat The raw data was collected from a 0.55T MR scanner (MAGNETOM Free.Max, Siemens Healthineers, Forchheim, Gemany) with a PD-weighted SPACE sequence (using DISORDER trajectory) from a 29-year-old male healthy volunteer under head and mandibular movements. The sequences will be open-sourced on Siemens' C2P platform soon. To use with the open-sourced code, please put the raw into a generated subfolder under /Studies-deploy (e.g., /Studies-deploy/1). Then modify the path in the main script 'batch_dental_multiple.m' to align with the raw path. For example: rootFolder = './Studies-deploy';studiesFile = fullfile('./Studies-deploy', 'studies.m');numCases = 1;caseList = [1]; Trained nnUNet models for teeth segmentation and head segmentation To run the segmentation-enabled workflow, you will need nnUNetv2 installed in a Python environment. Essential links: nnUNet repository: MIC-DKFZ/nnUNet nnUNetv2 installation and setup: official installation guide PyTorch installation: PyTorch local installation guide The current batch script assumes: a Conda environment name such as nnunetv2 nnUNetv2_predict is available in that environment nnUNet paths are configured via nnUNet_raw, nnUNet_preprocessed, and nnUNet_results Please download the two models, unzip, and place them under your local nnUNet_results directory. In batch_dental_multiple.m, these are currently set through the local variables CONDA, ENVNAME, and NNUNET_BASE. You will likely need to edit these paths for your system before running the workflow. For detailed instructions of using the dataset, please refer to: git repo: https://github.com/ZihanNing/dental_MRI_motion_correction Paper: Motion-Robust Dental MRI for Imaging of Paediatric Dental Trauma [to be released]