The CORT Dataset

The matRad release includes five image datasets of phantoms and patients. Consequently, everybody can start treatment planning out of the box without the need for their own data. Besides a simple cubic box phantom, matRad also features the four datasets that have been published as the CORT dataset.

The CORT dataset (common optimization for radiation therapy) is an open dataset intended to be used by researchers when developing and contrasting radiation treatment planning optimization algorithms. It is comprised of datasets for a prostate case, a liver case, a head and neck case, and a standard IMRT phantom (TG119). In matRad, these datasets are already imported and ready to be loaded as MATLAB native .mat files. To allow rapid prototyping and reasonable performance in educational settings, the CT images were downsampled at import.

If you wish to take a look at the original data, you can find it here. The dataset contains the original Digital Imaging and Communications in Medicine (DICOM) computed tomography (CT) scan, as well as the DICOM structure file. In addition, the dose-influence matrix from a variety of beam/couch angle pairs is provided for each case.

Besides the pre-configured patient datasets, matRad also features a DICOM import.