Development of a Simplified Intravascular Brachytherapy Treatment Planning System Using Monte-Carlo Derived Dose Data
Abstract
Purpose
The goal of this study is to develop a simple treatment planning system (TPS) to assess the dosimetry of a Y-90 source train-based treatment for intravascular brachytherapy (IVBT) using data from Monte-Carlo simulations.
Methods
Using a Monte-Carlo simulation that was already built in Geant4, an accurate dose distribution was obtained of an IVBT treatment using a source train of Y-90. A simple TPS was designed using Python to overlay the dose distribution onto the great vessel of an anonymized cardiac CT scan. Dose normalization and remapping were accomplished by utilizing the pydicom, SimpleITK, and numpy Python libraries. Dose Volume Histograms (DVHs) were created to assess the dose to the heart, vessel walls, and lungs. Moreover, to simulate the positional uncertainty of the catheter, dose distributions were calculated in Monte-Carlo after shifting the catheter by a specified amount laterally in the blood vessel. This was then used to calculate DVH’s for the great vessels and then graphed onto the same plot for dosimetric analysis.
Results
Other than the blood vessel wall around the catheter, the dose to the heart and lungs is orders of magnitude lower than the prescription dose. This is expected as the dose from beta sources falls quickly with distance. The dose to the vessel wall, however, fluctuates significantly (> 9%) from the prescription dose, depending on the shift of the catheter.
Conclusion
We have developed a simple TPS using data from Monte-Carlo simulations to assess the dosimetry of a Y-90 source train-based IVBT. The results reaffirm IVBT is a highly localized treatment with negligible radiation doses to heart and lungs. The positional uncertainty, however, introduces fluctuations to the dose distribution to the vessel wall.