ESTRO 2021 Abstract Book

S1363

ESTRO 2021

Patient-specific QAs were performed with global gamma index criteria of 3%/1mm.

Results The dose differences between MC model and measurements showed a systematic dose underestimation of the MC model, with an average of 1.5% over all configurations tested. Additionally, for depths deeper than 12cm, the difference could increase up to 3.0%. The use of MLC for lung treatments led to a reduction of the treatment time by a factor of 1.5 with a similar tumor coverage and dose to OAR compared to IRIS collimator plans. All patient specific QAs fulfilled the gamma criteria. Conclusion The MC model for MLC of Precision TPS is clinically acceptable with dose differences of less than ±2%, but still lacks accuracy compared to what could be expected from a dose calculation algorithm. Considering the supposed superior accuracy of the MC algorithm for dose calculation, dose differences of less than ±1%, especially in simple configuration, were expected. Therefore, if the MC model is to be used clinically, we would recommend its use only in case of heterogeneous conditions and for shallow-seated tumors. PO-1642 Feasibility of incorporating bead TLDs for in vivo dosimetry into current Radiotherapy CT protocols M. Masterson 1 , T. Knapton 2 , I. Davidson 3 , R. Maor 3 , S. Jafari 4 1 TRUEinvivo Ltd., Research , Surrey, United Kingdom; 2 TRUEinvivo Ltd., Research, Surrey, United Kingdom; 3 TRUEinvivo Ltd., Software, Surrey, United Kingdom; 4 TRUEinvivo Ltd, Medical Physics/Research , Surrey, United Kingdom Purpose or Objective To study the feasibility of scanning Silica Bead TLD arrays (DOSEmapper TM ) using typical Radiotherapy CT planning protocols for use in in vivo dosimetry. In particular, this study will evaluate the ability to view and detect the Silica Bead TLDs in current CT protocols and to assess the reproducibility of the position of the TLDs on the day of treatments. Materials and Methods Using an in-house 3D ABS printed phantom and a Rando phantom, 3 forms of DOSEmappers TM were scanned (1D-in the Urinary tract, 2D – for the surface measurements of chest wall and 3D – arranged on a rectum insert) using a Toshiba CT simulator scanner. The phantoms were scanned under a range of current clinical settings. Replicating the scan position of the DOSEmappers TM on the phantom was investigated using tattoos and structural landmarks. Preliminary investigations were carried out of a bespoke software programme to automate the identification of the bead TLDs and further to assign the absorbed dose of that bead TLD to the image on the patient CT scan. Additional assistance tools were incorporated into the DOSEmappers TM design, such as higher density marker beads in known positions to help with the overall identification of the beads and the repeatable positioning of the TLDs on the phantoms on the subsequent scans and treatment session. Results The arrangement of the TLD bead array on each DOSEmapper TM type was modified to improve the Silica Bead TLD detectability on a 120 keV and 2 mm slice CT scan, the current design includes 2 TLD beads per measuring point with the distance between each point ranging from 1 mm to 10 mm. By modifying the arrangement of the TLDs, they became detectable both by eye (figure 1) and by the bespoke software using standard CT protocols (figure 2). The addition of higher density beads (gold) allowed for easier identification of the start and end points of the DOSEmappers TM .