(Tuesday, 7/31/2018) 9:30 AM - 10:00 AM
Room: Exhibit Hall | Forum 6
Purpose: Radiotherapy procedures offer a unique way to study the health effects of radiation because patients receive controlled and well-documented exposures. The key objective of such studies is to connect patient-specific organ dose with the risk of late-term morbidities. Unfortunately for many retrospective epidemiological studies, it is not possible to access anatomical computed tomography (CT) data for performing the dosimetry. In such cases, a whole-body computational human body phantom can be used as a surrogate for the unknown patient anatomy. The purpose of the present work is to connect computational phantom resources to commercial radiotherapy treatment planning systems (TPS) for the estimation of radiotherapy organ dose.
Methods: A MATLAB software tool, called the DICOM-RT Generator, was developed to convert computational voxel phantoms into a Digital Imaging and Communications in Medicine radiotherapy (DICOM-RT) format. A DICOM CT image set of the phantom is created through a density-to-Hounsfield unit calibration curve. An accompanying structure set (DICOM-RTSTRUCT) is created by generating a binary mask of each significant organ and tracing the boundaries on each CT image slice.
Results: The software, including a user-friendly graphical user-interface, was developed and tested on a library of computational phantoms. The resulting DICOM-RT files were tested on commercial TPS (including Eclipse and Pinnacle) and several widely-used DICOM viewers. Examples will be discussed using the DICOM-RT Generator to estimate organ dose for members of the National Wilms Tumor Study cohort.
Conclusion: Through generation of DICOM-RT files for surrogate computational phantoms, radiotherapy plans may be added within TPS to estimate dose received by a patient with unknown anatomy. Auto-generation of the contours can be significantly helpful for organs that are difficult to segment, such as the colon and red bone marrow. A standalone MATLAB-compiled software package of the DICOM-RT Generator will be made available for others to use.
Funding Support, Disclosures, and Conflict of Interest: This work was funded by the intramural program of the National Institutes of Health, National Cancer Institute, Division of Cancer Epidemiology and Genetics. The use of MATLAB within this abstract does not necessarily imply recommendation by the National Institutes of Health.
Keywords
Contour Extraction, DICOM-RT, Phantoms
Taxonomy
Contact Email
