Maryam Khatonabadi, Presenter: Nothing to Disclose

Di Zhang PhD, Abstract Co-Author: Employee, Toshiba Corporation

Christopher H. Cagnon PhD, Abstract Co-Author: Nothing to Disclose

John J. Demarco PhD, Abstract Co-Author: Nothing to Disclose

Michael F. McNitt-Gray PhD, Abstract Co-Author: Institutional research agreement, Siemens AG Research Grant, Siemens AG Instructor, Medical Technology Management Institute

The purpose of this study was to explore water equivalent diameter as a size metric in estimating organ doses and to compare its performance with equivalent diameter and perimeter.

Voxelized models were created from adult patients undergoing abdomen/pelvis (39) and thoracic (32) CT exams. For abdomen/pelvic scans, the liver, spleen and kidneys were segmented and explicitly represented in voxelized models; similarly for lungs and glandular breast tissues in thoracic scans. TCM functions were extracted from patients’ projection data and used in Monte Carlo simulations, which explicitly modeled the scanner and patient to obtain estimates of organ dose. Additionally, the whole body was segmented from surrounding air and table using semi-automated methods. This was performed on each patient at a specific anatomical location to calculate: (a) patient perimeter, (b) equivalent diameter and (c) water equivalent diameter (Dw) of the segmented area, taking into account body attenuation. In addition, for each patient, an approximated water equivalent diameter (ADw) was also calculated, which utilized the entire axial image (i.e. not just segmented region). Linear regression analysis was used to investigate the relationship between each size metric and organ doses.

For liver, ADw showed a better correlation (R2=0.86) compared to perimeter and equiv. diameter (R2=0.65). Similarly for lungs, ADw resulted in a better correlation (R2=0.45) while perimeter and equiv. diameter showed lower correlation of 0.34 and 0.37, respectively. The low correlations are attributed to a noticeable separation of the lung organ dose by gender, specifically for the female chest models. A correlation of 0.9 between lung dose and ADw was observed when male patient models were analyzed separately.

Overall Dw and ADw seem to correlate better with organ dose compared to equiv. diameter and perimeter. The segregation between males and females in lung dose from thoracic scans may be due to anatomical and composition differences between genders in the thoracic region; these differences remained even when Dw and ADw were used. Further study of size/attenuation metrics is required here.

Both Dw and ADw correlate well with organ dose and can be calculated on per image basis and potentially reported in DICOM as a patient size for use in future organ dose estimation methods.

Khatonabadi, M, Zhang, D, Cagnon, C, Demarco, J, McNitt-Gray, M, Water Equivalent Diameter (Dw) as a Patient Size Metric for Estimating Organ Dose to Patients Undergoing CT Exams.  Radiological Society of North America 2012 Scientific Assembly and Annual Meeting, November 25 - November 30, 2012 ,Chicago IL. http://archive.rsna.org/2012/12034348.html