ESTRO 2023 - Abstract Book

S1583

Digital Posters

ESTRO 2023

will give the confidence required to reduce target volumes by applying personalised margins based on variability due to breathing.

PO-1846 Characterization of a Novel High Performance kV CBCT System on a Ring-Gantry Linac

M. Salerno 1 , A. Haertter 1 , B. Koger 1 , C. Kennedy 2 , B.K. Teo 1 , M. Alonso-Basanta 1 , L. Dong 1 , T. Li 1

1 University of Pennsylvania, Radiation Oncology, Philadelphia, USA; 2 University of Pennsylvania, Radiation Oncology, Philadephia, USA Purpose or Objective The Varian HyperSight™ CBCT system features a new large area 86cm x 43cm imager that enables 6 second acquisition and advanced model-based scatter reduction added to reconstruction. This study assessed the image quality performance of this novel CBCT system by comparing to existing CBCT systems on TrueBeam, Halcyon version 2.0, and the Siemens Somatom Edge CT simulator. Materials and Methods Catphan 504 was scanned using the HyperSight imager on an upgraded Halcyon linac with Head, Breast, Thorax Slow, Pelvis and Pelvis Large protocols designed for planning on CBCT (CBCTp). Images were reconstructed using the iCBCT Acuros algorithm with varying slice thickness (1, 2, 3, and 5 mm). Automated image analysis was performed with Pylinac (//github.com/jrkerns/pylinac). Analysis included tests for HU linearity, slice thickness, and relative modulation transfer function (MTF). Low contrast visibility was also assessed with the CTP515 module. The visibility score for each ROI (I) is given by: Visibility(I)=Contrast(I) ×√ (Area(I) × DQE(I)) Results HU linearity results for the tested CBCTp protocols are shown in Table 1. Measured HU values were found to be within the Catphan 504 nominal range for all materials except for Delrin and Teflon (within 10 HU overall but similar to diagnostic CT). Figure 1 showed the nominal CT reconstructed slice thickness vs measured slice thickness, and demonstrated excellent agreement, within 0.25mm for the 1-2mm selected width and 0.5mm for the 3-5mm selected width. The calculated relative MTF for each of the tested protocols is shown in Figure 2, which is comparable to the Siemens Somatom Edge CT simulator in our clinic. Low contrast visibility scores for each protocol (2mm slice thickness for Head, Breast, Thorax Slow, Pelvis and Pelvis Large) were 13.41, 10.74, 12.64, 12.21 and 18.32, respectively, substantially improved from scores of current CBCTs on TrueBeam (10.2) and Halcyon (8.39) when using Pelvis protocol and 2mm slice thickness.

Head Breast Pelvis Pelvis Large TrueBeam Halcyon Somotom Edge Nominal

Air

-991 -993 -992 -991 -187 -189 -186 -184 -100 -101 -98 -97.5 108 111 113 111 334 335 338 333 939 939 939 931.5 -48 -48

-1000

-1000 -1022

-1016

PMP

-188

-215 -111

-190.5

-196 -104

LDPE

-98 -38 130 362 975

-96 -37 126 345 950

Polystyrene -49 -49

-68

-47

Acrylic Delrin Teflon

97

114.5 365.5

323 899

1000.5 Table 1: HU linearity, measured for the HyperSight Head, Breast, Pelvis, and Pelvis Large CBCTp protocols with the Catphan 504 phantom. Images were reconstructed with the iCBCT Acuros algorithm and 2mm slice thickness. Results were compared against TrueBeam, Halcyon 2.0 Pelvis CBCT protocols with 2mm slice thickness, Siemens Somatom Edge Brain protocol with 3mm slice thickness, and Nominal HU values.