An EGS4-ready tomographic computational model of a 14-year-old female torso for calculating organ doses from CT examinations

Abstract

Fifty-four consecutive CT scans have been used to construct a tomographic computational model of a 14-year-old female torso suitable for the determination of organ doses from CT. The model, known as ADELAIDE, is in the form of an input file compatible with user codes based on XYZDOS.MOR from the readily available EGS4 Monte Carlo radiation transport code. ADELAIDE's dimensions are close to the Australian averages for her age so the model is representative of a 14-year-old girl. The realistic anatomy in the model differs considerably from that in Cristy's 15-year-old mathematical computational model by having realistically shaped organs that are appropriately located within a real external contour. Average absorbed dose to organs from simulated CT examinations of the chest and abdomen have been calculated for ADELAIDE using EGS4 within a geometry specific to the General Electric Hi-Speed Advantage CT scanner and using an x-ray spectrum calculated using data from the scanner's x-ray tube. The simulations include the scanner's beam shaping filter and patient table. It is suggested that the resulting values have fewer possible sources of uncertainty than organ doses derived from dose coefficients calculated for a MIRD style model with mathematical anatomy and a spectrum that may not match that of the scanner. The organ doses were normalized using the scanner's CTDI measured free-in-air and an EGS4 simulation of the CTDI measurement. Effective dose to the torso from 26-slice chest and 24-slice abdomen examinations (at 120 kV, 200 mAs, 7 mm slices) is 4.6 +/- 0.1 mSv and 4.3 +/- 0.1 mSv respectively.

Patient‐specific CT dosimetry calculation: a feasibility study

“…Caon et al ( 61 ) reported on a CT dosimetry study for a 14‐year‐old female tomographic phantom. The EGS4 radiation transport code was used for Monte Carlo simulation of a GE Hi‐Speed Advantage CT scanner operating at 120 kVp with 10 mm beam width.…”

Multi-Organ Segmentation of CT Images using Statistical Region Merging

“…The images have a pixel size of 2.53 × 2.53mm and slice separation of 10mm. The data set is referred to as ADELAIDE [1].…”

“…Models of human anatomy are required in order to allow the calculation of organ doses and effective dose received from diagnostic radiography procedures such as computed tomography [1]. The cross-sectional (tomographical) medical images produced by medical imaging are a source of data for constructing models of human anatomy.…”

Feasibility study of a new platform based on the Case-Based Reasoning principles to efficiently search and store voxel phantoms

“…More phantoms were developed like the NORMAN one (for NORmal MAN) of the National Radiological Protection Board (Dimbylow, 1998), and the one of Zubal (Zubal et al, 1994). ADELAIDE (Caon et al, 1999), GOLEM (Zankl et al, 2001), developed by Zankl and Wittmann are also voxel phantoms, same for FAX (Kramer et al, 2003) and MAX (Kramer et al, 2004) that were made to fit the ICRP Publication 89 (ICRP, 2002) recommendations regarding organs weights. VIP-Man created by Xu (Xu et al, 2000) and the Canine Anatomic Phantom designed by Padilla (Padilla et al, 2008).…”