Shared data for intensity modulated radiation therapy (IMRT) optimization research: the CORT dataset.

Area of Science:

• Medical Physics
• Radiation Oncology
• Computational Biology

Background:

• Developing and contrasting radiation treatment planning optimization algorithms requires standardized datasets.
• The CORT dataset (common optimization for radiation therapy) is introduced to facilitate algorithm development.
• Existing methods for radiation therapy planning optimization lack a common benchmark for comparison.

Purpose of the Study:

• To provide a common dataset for researchers developing and comparing radiation therapy treatment planning optimization algorithms.
• To enable one-to-one comparisons of different optimization algorithms for intensity modulated radiation therapy (IMRT).
• To support research in beam angle optimization, volumetric modulated arc therapy, and direct aperture optimization.

Main Methods:

• The CORT dataset includes common datasets for various treatment planning scenarios.
• Datasets encompass prostate, liver, head and neck cancer cases, and a standard IMRT phantom.
• Dose-influence matrices, DICOM CT scans, and DICOM structure files are provided for each case.

Main Results:

• The CORT dataset provides essential dose-influence matrices, detailing dose to patient voxels from each pencil beam.
• Included are original DICOM computed tomography (CT) scans and DICOM structure files for each case.
• The dataset facilitates the optimization process by providing the primary entity needed for calculations.

Conclusions:

• An open dataset, the first of its kind, is now available to the radiation oncology community.
• This resource will enable researchers to compare and validate methods for optimizing radiation dose delivery.
• The CORT dataset promotes reproducible research and accelerates advancements in radiation therapy planning.