Evaluation of image quality in children chest CT using adaptive statistical iterative reconstruction

Objective The purpose of this study was to compare radiation dose,noise and image quality of low-dose chest CT reconstructed with adaptive statistical iterative reconstruction(ASIR)and routine-dose CT reconstructed with filtered back projection(FBP).Methods Eighty children(study group)(age 29d~16y,mean age 3.3years)underwent low-dose chest CT with 30% ASIR.Age-dependent noise index(NI)was used:NI=11for 0~12monthes old,NI=13 for 1~3years old,NI=15for 3~6years old,and NI=17for 7~17years old.A group of 80children(control group) underwent routine-dose chest CT with FBP.NI for the control group was set to 9HU for 1~7years old and 12HU for 7 ~17years old.Two radiologists independently evaluated images for abnormal CT findings,normal lung structures,and image noise on a 5-point scale with 3being clinically acceptable.Quantitative measurements of radiation dose and image noise on left ventricle and muscle were also obtained and statistically compared between the two groups.The inter-observer agreement was tested with Kappa statistics.Results The image noise for the study group was 10.49±1.65,slightly higher than 9.16±1.52for the control group,with no significant difference(P0.05).However,the mean image quality scores for the study group and control group were 3.89and 3.90,respectively.All images were judged to be acceptable for clinical diagnosis,and there was good inter-observer agreement in diagnostic acceptability(k=0.64).On the other hand,the average CTDIvol for the study and control groups were(1.19±0.78)mGy and(2.62±0.93)mGy,respectively.The 30% ASIR provided 54.6%lower dose than FBP with significant difference(P0.001).Conclusion Compareing with routine-dose CT with FBP,low-dose CT with 30% ASIR provided 54.6% dose reduction in pediatric chest CT examinations,while maintaining diagnostically acceptable images.