Keywords:
Dosimetric comparison, Radiation safety, Dosimetry, CT, Radioprotection / Radiation dose, Radiation physics
Authors:
E. Lilla, E. Balázs, C. Sokvári, Z. Dankó, L. Urbán, P. Bágyi, L. Balkay; Debrecen/HU
DOI:
10.26044/ecr2019/C-2337
Results
The ROI noise indexes (SD and SD/mean) and EffDose were inversely changed across the scanners (Fig.1 and Fig.2).
However there were fewer differences to notice between the GE scanners,
the noise indexes were practically same and the median EffDose values were 27.8 and 28.8 mSv.
Fig.
3-5 show the detailed correlation plots between standard deviation and the effective dose values for each CT study and in the case of the three CT scanners.
The greater EffDose spread is characterized by CT3,
while the smaller is by CT2.
In all three cases the fat SD distribution is the narrowest and the SDs of the fat tissue were the smallest (< 0.18).
In addition,
the aorta SDs had the highest deviation across the CT scanners.
Furthermore,
comparing the CT1 and CT2 systems (Fig.
3 and Fig.
4),
the maximum value of the EffDose is about 50 and 60 mSv,
for CT2 and CT1,
respectively.
This can be explained by the fact that the maximal current in the AEC settings of CT2 is lower (Table 1.). We also investigated how the patient weight affects the image noise and the resulted dose saving.
We defined three weight groups (0kg<m<70 kg; 70kg<m<100kg; m> 100kg) and plotted the median CV and EffDose values across the groups and scanners.
Fig.
6 shows that the predefined AEC settings produced similar image noise up to 100 kg body weight,
but higher noise was observed in greater weights at all CT systems.
It was also observable that the tissue CVs generally increased with the body mass,
and when excluding the fat tissue almost constant CV-s were calculated.
The aorta areas showed the largest CV values that can be due to the proximity to the spine,
which may generate artifact to the vessel compartment.
Irrespective of the tissue type,
the smallest CV is characterized by the CT3 system.
As a next step,
the FOM values were analyzed among the scanners.
In the CT field the 1/(SD^2*EffDose) parameter is frequently used thus we plotted this FOM value for the different CT scanner for each tissue type.
Fig.
7 presents the result and it is clearly visible that CT3 has the largest FOM independently of the tissue type.
Another interesting fact is that the FOM distribution is very similar for the CT1 and CT2 systems.
These belong to the same vendor (GE),
but the technical parameters (number of detector slice,
collimation,
X-ray tube,
AEC options,
etc.) were not the same.
The main reason for this FOM similarity is the built-in detector,
because the included detector materials and primary pixel sizes are same for both scanner (as we were informed by the local GE representative).
In addition,
the AEC settings were adjusted by two different medical teams (but with same expectations),
thus the AEC settings are not exactly the same for the CT1 and CT2 system.
Nevertheless,
the AEC settings and the resulting dose saving can be similar because the image quality expectations were similar in our optimization teams and because the AEC algorithms came from the same manufacturer.
Since CT3 system has the highest FOM and is characterized by the highest EffDose currently,
the three systems could be harmonized in terms of image quality by increasing the dose saving of CT3.
In the future we analyze and investigate the optimum AEC setting of CT3 system to allow for the similar image quality of the three cameras.