Study of dosimetric quantities and image quality in pediatric examinations of chest and abdomen computed tomography

This work had the objective to achieve the knowledge of the dosimetric quantities related to chest and abdomen computed tomography (CT) examinations of pediatric patients, in Belo Horizonte city. The reason of this work is based on the fact that the probability of health detriment in children, which it may be caused by radiation, is higher than in adults. Besides, although in many countries the knowledge and control of patient doses is a normal procedure, this safety culture does not exist in Brazil. Another objective of this work was to compare the dosimetric quantity values with the Diagnostic Reference Levels (DRLs); when it was needed, an optimization process was applied and the quality of the diagnostic image obtained with the optimized technical parameters was analyzed. This study was carried out in five hospitals, where the weighted air kerma index (Cw), the volumetric air kerma index (Cvol), the air kerma - length product (PKL,CT), the Effective Dose (E) and the Normalized Effective Dose (En) were determined; three methods were adopted for measurements: the ionization chamber inside a chest pediatric phantom, radiochromic films and the CT-EXPO software. The optimization process was applied to a single hospital through variations in the current (mA) and voltage (kV) of the x-ray tube for the protocols used for abdomen CT examinations. The analysis of the quality of the diagnostic image was done by Normal Distribution and ROC analysis; spatial resolution analysis was used through MTF determination and the noise level was judged in terms quantitative and qualitative. Results of the dosimetric quantities showed that they significantly differed between single-slice and multi-slice tomography units, but their values were always below the recommended DRLs. The optimized values of the dosimetric quantities obtained after the optimization process showed that it was possible to reduce the radiation exposure of pediatric patient without losing the image quality, despite the low dose values found. Dose values very lower than DRLs might induce to an unjustified practice of increasing the dose for improving the image diagnostic quality; the results showed that if it was done there was not a relevant change in the quality. Based on the results of this work, it was possible to state that the dosimetric quantity values in the studied pediatric tomographic procedures were lower than DRLs, but there was still room of optimization for patient protection and to maintain an adequate image diagnostic quality in computed tomography. (author)