K. Mitsui et al.: Radioprotection 2025, 60( 1), 37 – 43 41 Fig. 6. Relationships of CTDIvol( a) and SSDE( b) against BMI. The solid lines represent regression lines.
imaging sections. The SSDE is an index of the patient dose that considers tissue attenuation, and is calculated by multiplying the CTDIvol by the f factor determined based on the WED. As the head size increases, the f factor decreases; therefore, the SSDE increases less than the CTDIvol. However, it was demonstrated that the SSDE still increased with increasing ED or WED, suggesting that a larger head receives higher absorbed dose in brain CT. Therefore, it is desirable to consider the head size in radiation dose management using the SSDE.
The DLP is the integral of the CTDIvol over the longitudinal scan range. The plots of the DLP against the ED and WED appeared to be divided into two groups, and the overall correlation of DLP with ED or WED was relatively low. In routine radiation dose management, CT scans with exceptionally high or low radiation doses are identified, and the cause of each outlier is investigated. There are larger variations in DLP at a given head size than in CTDIvol, and the sensitivity for identifying outliers caused by technical problems other than inappropriate setting of the scan range would be lower for the DLP. Assessment of the CTDIvol appears to be indispensible in identifying outliers in brain CT, and that of the DLP would be supplementary. In this study, we performed CT in the axial mode but not helical mode, with a beam width of 2 cm. As a result, the scan length varied not continuously but in 2-cm increments, which is responsible for the division of the plots into two groups. The use of the helical mode may improve the correlation of the DLP with the ED or WED.
A high correlation between radiation dose indices and body weight has been reported for CT of the body( Israel et al., 2010; Iball and Tout, 2014; Inoue et al., 2016, 2022d). In this study, the CTDIvol and SSDE in brain CT showed positive correlations with body size indices, including body weight, body height and BMI. Although the correlations were highest for weight among them, they were relatively low, presumably reflecting the weak relationship between the head size and body weight. Considering the body size appears to be less significant in brain CT than body CT. The ED and WED are preferred to body weight to identify examinations that delivered exceptionally high radiation doses. However, the WED and ED are not always readily available. When they are unavailable, considering body weight may be of some significance for radiation dose management.
There are limitations to this study. Only one CT scanner was used, and radiation dose was modulated using inbuilt AEC software. The type of AEC software and parameter settings of the software affect dose modulation( Inoue, 2022a); therefore, they would influence the relationship between the head size and dose index. Studies using different methods of dose modulation remain to be done. Moreover, optimization of radiological imaging involves reducing radiation dose while maintaining diagnostic performance. In this study, we showed that radiation dose indices reflect the WED when using AEC; however, we did not evaluate image quality and did not show whether optimization was actually achieved.
5 Conclusion
When adult brain CT was performed using AEC, the CTDIvol showed high positive correlations with the ED and WED, suggesting that AEC achieved dose modulation reflecting X-ray attenuation in the head. It is recommended to evaluate the relationship of the CTDIvol with the ED or WED for radiation dose management of adult brain CT. As the ED and WED increased, the SSDE and DLP also increased. However, the correlation with the ED and WED was relatively weak for the DLP, presumably attributable to imaging in the axial modes. The CTDIvol and SSDE showed weak positive correlations with body weight, and considering body weight may have some significance in dose management for brain CT.
Funding
This research did not receive any specific funding. Conflicts of interest
The authors declare no conflict of interest.