A Comparative Study of Prospective Scanner-Reported and Retrospective Radiation Dose Index Monitoring (RDIM) Software Size-Specific Dose Estimates (SSDE) for Abdomen-Pelvis CT
Abstract
Purpose
The AAPM TG Reports 204 and 220 size specific dose estimate (SSDE) provides a more useful CT dose approximation compared to CTDIvol. Size-Specific Dose Estimate (SSDE) corrects for patient size and is commonly used in radiation dose index monitoring (RDIM) applications to assess the appropriateness of CT exam doses. New CT software has emerged with prospective console displays of series-level SSDE, calculated using the patient scout image(s). This study compares SSDE values from CT scanner displays and the RDIM application (Imalogix) to determine their consistency for both prospective dose assessment by the technologist and retrospective data review.
Methods
SSDE data was extracted from 384 abdomen/pelvis patient exams on two CT platforms (Siemens X.Cite, Canon Aquilion One) by accessing the RDSR and Imalogix reports. Exam data was sorted by size and evaluated collectively and within patient-size categories. SSDE data was evaluated using the Wilcoxon Signed Rank Test and visualized for agreement with Bland-Altman plots.
Results
Wilcoxon Signed Rank Test determined that SSDE datasets were statistically different from each other (p<0.05), with a median difference of 3.9% and the most dramatic deviations (<22%) observed for larger patient sizes. Bland-Altman plots demonstrate reasonable agreement for all groups of SSDE (average differences <2mGy), with agreement degrading with increasing patient size. Across all groups, scanner reported SSDE is generally lower than the RDIM reported SSDE.
Conclusion
Although differences in patient size, the scanner reported SSDE is typically lower than the RDIM reported SSDE, making it feasible to use the scanner reported SSDE prospectively by technologists to assess the appropriateness of an exam’s dose. Future work may include comparing reported SSDE with additional scanners or exam types, and utilizing phantom images to evaluate scanner and RDIM reported SSDE against image-based SSDE calculations.