Validation of absorbed-dose calculations using Y-90 in different source and target geometries

Because of increasing interest in patient-image-based dosimetry, dosimetry validation in phantoms would be useful, particularly for modeling complex source-target geometries. We have devised a fluorescence technique to measure doses to liquid-filled compartments. Solutions containing 0.1 mM coumarin-3-carbosylic acid (CCA) with Y-90 (YA) and without Y-90 (NA) were prepared. We measured the self-absorbed dose in six spheres, three cylinders, and two shell-sphere geometries. Each shell-sphere geometry consisted of an inner sphere containing NA solution, surrounded by an outer shell containing YA solution. This enabled us to measure the self-dose to the source shell as well as the cross-dose to the target sphere. After a 24 or 48 hour exposure, the mean dose to each compartment was estimated by measurement of the induced fluorescence in the solutions. A calibration curve obtained using NA solution exposed in a Cs-137 irradiator (98.3rads/min) for different times displayed a linear increase in fluorescence with increasing dose up to 1500rads. The self-dose increased with sphere or cylinder size, and agreed within ±8.5% with values computed using a Y-90 dose point-kernel method. However, the dose delivered to the 'cold' sphere compartment from the adjacent 'hot' shell, which was a few percent of the self-dose to the hot compartment, showed a large discrepancy between measured and calculated values. CCA fluorescence dosimetry is a promising technique for verification of dose calculations.