Akademik Veri Yönetim Sistemi
Applied Radiation and Isotopes, cilt.232, 2026 (SCI-Expanded, Scopus)
The increasing use of computed tomography (CT) in pediatric patients has prompted researchers to investigate the radiation dose associated with pediatric CT scans. Studies optimizing image quality and minimizing radiation dose in pediatric CT examinations continue to increase each year. This study aimed to determine the radiation doses received by pediatric patients during CT scans of the head, thorax, and abdomen, and to evaluate the image quality of pediatric thorax CT images using different iterative reconstruction (IR) algorithms. The study was conducted on CT systems of various brands and models installed in three major university hospitals in Ankara. Radiation doses for pediatric head, thorax, and abdomen CT examinations were measured in volume computed tomography dose index (CTDIvol), size-specific dose estimate (SSDE), and Dose length product (DLP), and local diagnostic reference level (DRL) was determined for each examination. Image quality was evaluated from thorax CT images of children aged 4-6 years, objectively and clinically, depending on different IR algorithms (iDose4, advanced modeled iterative reconstruction (ADMIRE), and adaptive statistical iterative reconstruction (ASiR)). Median CTDIvol/DLP values for pediatric head/thorax/abdomen CT were 31.6 mGy/541.3 mGy cm, 2.1 mGy/30.1 mGy cm, and 2.2 mGy/39.8 mGy cm, respectively. The median SSDE (mGy) calculated according to the American Association of Physicists in Medicine (AAPM) 220 report for pediatric thorax and abdomen CT scans were 4.7 and 4.6 mGy. The noise decreased with increasing iteration level for each IR algorithm, while the SNR and CNR increased. Absolute visual grading analysis (VGAabs), which gives clinical image quality, did not increase or even decrease with increasing IR level, except for one CT system. The interobserver agreement was poor (ICC = 0.29, p = 0.30), moderate (ICC = 0.72, p = 0.04), and good (ICC = 0.84, p = 0.02) for images created with iDose4, ADMIRE and ASiR, respectively. While the objective image quality improves with each IR algorithm's increasing iteration level, no significant change in clinical diagnostic quality was observed.