Comparison between manual-calculated and IndoseCT-calculated SSDE based on Deff and Dw methods on truncated CT images

Abstract

IntroductionSize-specific dose estimation (SSDE) is suggested for accurate measurement of CT dose. Various methods have been developed for SSDE calculation by both manual and automated methods. However, truncation artefact may affect the specific size determination and SSDE calculation. In this article, we evaluate how truncation affects measurements of SSDE and compare the manual SSDE measurement based on individual sizes (effective diameter, Deff and water equivalent diameter, DW) with the automated calculation using IndoseCT software on truncated CT images. MethodsA phantom study was conducted using Siemens SOMATOM Definition AS + CT scanner with two CTDI phantoms of different sizes, 22 cm and 32 cm. Phantom images were acquired by varying the phantom positioning to simulate truncation percentages (TP) of 5%, 10%, 15%, and 20%. The SSDE values were manually calculated based on the Deff and Dw methods (AAPM, 2014). SSDE was also calculated automatically using IndoseCT software (Version 20 b) and compared with manual calculation. The percentage error due to different TP was calculated in comparison to the non-truncated image (TP 0%). The errors due to truncation on SSDE calculation was also compared between both manual and automated methods. Statistical analysis was carried out to determine the statistical differences and correlation. ResultsThe findings show that Deff for non-truncated image yields better results with lower percentage errors (mean PEmanual = 0.75% and mean PEautomated = 0.14%) as compared to Dw (mean PEmanual = 6.11% and mean PEautomated = 5.81%). There is a positive strong correlation between percentage error and truncation percentage. There are strong positive correlations between TP and percentage error, PE (r = 0.96 and r = 0.94) in determination of SSDEDeff and SSDEDw, respectively. Deff approach resulted higher percentage error as compared to Dw. ConclusionTruncation is strongly correlated with SSDE calculation, and therefore correction for truncation artefact is crucial to minimise error in SSDE measurement. The automated calculation of SSDE can be confidently replaced the manual method in both Deff and Dw approaches to ensure rapid and accurate SSDE measurement.