This parquet file can be parsed by Python. It contains 37 columns of data related to the daily collection of voxelwise quantitative MRI parameters from a cohort of subjects being treated for head and neck squamous cell cancer with primary radiotherapy. The columns are as follows: X, Y, Z, XYZ: coordinate columns reflecting the DICOM positions of each voxel Value ##: raw values from the DICOM files Smoothed Value ##: values after a 2mm radial smoothing kernel was applied patient_id: subject the data was pulled from contour: the ROI the value was pulled from - control or GTVp (gross primary tumor volume) parameter: qMRI parameter evaluated