H. Murat et al.: Radioprotection 2025, 60( 1), 65 – 75 67
longer scan times or higher radiation doses( Bouchareb et al., 2023).
E ¼ k⋅DLP: ð3Þ
2.2 Acquisition protocols
All examinations were conducted on PET / CT system( DST Discovery, GE, USA) that was equipped with BGO crystalbased detector. The scanner was calibrated in accordance with the guidelines for clinical and research studies. Patients were required to fast for a minimum of 6 hours prior to the injection of 18 F-FDG, and their glucose levels had to be lower than 10 mmol / L( 180 mg / dL). The mean amount of 18 F-FDG injected was 443.8 MBq, ranging from 200.7 to 646.3 MBq. Patients were instructed to empty their bladder and occupy the uptake room before the PET / CT examination, which was usually conducted in three-dimensional mode for increased sensitivity. After CT acquisition with an automatic tube current modulation( AEC) protocol, PET images were acquired at three minutes per bed position to match the same anatomical extent as the CT images. A low-dose CT scan was often performed for PET attenuation correction and anatomical localization( Boellaard et al., 2015; Koopman et al., 2016)
2.3 PET / CT Dosimetry
In this study, we employed IDAC-Dose 2.1, a Monte-Carlo based software, to campute both the absorbed and effective dose linked to PET imaging. To perform PET dosimetry, we entered the specific activity of the radiopharmaceuticals used, as well as patient-specific parameters in order to accurately model the distribution of radiation dose in different organs. The input data also necessitates the cumulated activity, Ã in a source region r s, relative to the administered activity as outlined in Eq.( 1)( Andersson et al., 2017b).
~ A
A o
¼ ∫ TD
0 Aðr s; tÞ · dt
A o
; ð1Þ
where A( r s, t) is the activity of the radiopharmaceutical in source region r s at time t, and A o is the administered activity. This cumulated activity per administered activity for a biokinetic model of 18 F-FDG provided by ICRP Publication 128( ICRP, 2015).
In the meantime, IDAC-Dose 2.1 was used to compute the effective dose, E for PET by summing up these organ doses and applying tissue weighting factors according to the latest ICRP publications, as outlined in equation( 2)( ICRP, 2008, 2021).
E ¼ X T w TD T; ð2Þ
where w T is the tissue weighting factor, D T is organ absorbed dose.
In order to calculate the E from the CT component, we first obtained the dose length product( DLP) values from the CT control console. Subsequently, the contribution of E from CT was calculated using the following equation( 3). k is a conversion factor that is based on the body part being scanned and the patient age at exposure as stipulated in ICRP Publication 103, and AAPM Report No. 96, respectively and it is expressed in mSv mGy �1 cm �1( AAPM, 2008; ICRP, 2007).
2.4 Organ cancer risk based on LAR of cancer incidence
The organ doses were converted into organ cancer risks by utilizing patient age at exposure and gender-specific data from BEIR VII phase II report( Huda et al., 2011; ICRP, 2022; Karim MKA et al., 2016; National Research Council, 2006). The main process involved tabulating the b coefficients based on Table 12 D-1 in the report through a 3 rd order polynomial fitting. Subsequently, the LAR was calculated based on b coefficients and the age of exposure for specific organs and gender as given in equation( 4)( Halid et al., 2018; Huda & He, 2012).
LAR ¼ b 0 þ b 1 Y þ b 2 Y 2 þ b 3 Y 3; ð4Þ
where b is the curve fit coefficient, and Y is the patient’ s age at exposure.
The estimation of organ cancer risk was determined by multiplying the organ dose and adjusted LAR according to age at exposure and gender, as listed in Table 1. To ensure that the organ cancer risk values correspond to the actual organ dose, we adjusted the organ dose based on the actual dose received compared to the standard dose of 0.1 Gy. Therefore, the organ doses for each patient were converted into organ cancer risks by utilizing the given equation( 5). An independent t-test was conducted to test any differences that existed between two groups. The summary elements for predicting the LAR of cancer incidence are shown in Figure 1.
Organ cancer risk ¼ D T LAR; ð5Þ
0:1
where D T organ absorbed dose by specific organ calculated in unit Gy, and LAR is the probability per unit dose( Gy) of developing cancer in that specific organ over a lifetime.
2.5 Statistical analysis and AI tools
The data obtained was recorded and analyzed using Excel spreadsheet with XrealStats plugin. In this study, the data variables were expressed as mean values ± standard deviation. The normality of the data was assessed using the Shapiro – Wilk test, and differences were tested using t-test with significance level of < 0.05. The LAR in the selected radiosensitive organs was then analyzed using the Kruskal – Wallis test. The test compared the mean ranks of LAR values across the different organs to determine if at least one organ’ s LAR was significantly different from the others.