SU‐FF‐T‐236: Functional Representation of Tissue Phantom Ratios for Small Photon Fields

• Published in: Medical physics (Lancaster) Vol. 34; no. 6; p. 2456
• Main Authors: Sauer, O A, Wilbert, J
• Format: Journal Article
• Language: English
• Published: American Association of Physicists in Medicine 01-06-2007
• Subjects: Data analysis; Field size; Medical treatment planning; Modulators; Photon scattering; Photons; Signal generators; Silicon detectors; Surface hardening; Water quality
• ISSN: 0094-2405; 2473-4209
• DOI: 10.1118/1.2760897

Purpose: For small photon fields tabulated TPR data are not readily available. Aim of this work was to find a functional representation valid at all depths d and for fields as small as 4 mm side length s. Such functions generate high quality beam data as input to a treatment planning system, or for calculating monitor units and dose to any point in water for the full range of s and d required clinically. Method and Materials: For beams of 6 and 10 MV collimated with the Elekta Beam Modulator, TPR was measured in water with a shielded silicon detector. Field sizes were from 4×4mm2 to 160×210mm2. For each field size the TPR was fitted to
TPR
 = (
D
s
 + (1 − 
D
s
)(1 − β
d
)) α 
exp
(− μ (1 − η 
d
) 
d
)
. The first factor describes the build up, with Ds being the surface dose and β a build up gradient. α is a normalization factor, introduced to normalize TPRs to 1 at a depth of 10 cm. The exponential, describes the declining part of the curve, where μ is a pseudo attenuation coefficient and η a beam hardening coefficient. The field size dependence of the parameters was analyzed separately and appropriate functions were fitted to these data. The surface dose Ds was set to the respective mean values, while β(s) was fitted to a power function and μ(s) and η(s) to exponentials. Results: The measured TPR data could be fitted very well with the proposed function. Owing to the increasing amount of scattered photons μ decreases with field size. The beam hardening η turned out to be negative except for field sizes below 1 cm. Conclusion: The proposed model predicted TPRs generally better than with 1 % accuracy at all field sizes and depths. Its implementation improved the quality of monitor unit calculations.