Abstract
Computed tomography (CT) is an important imaging modality that provides cross-sectional images, aiding in the detailed visualization of internal structures for accurate diagnosis and treatment. The pediatric population is more sensitive to radiation than adults, making radiation dose (RD) optimization an important concern in pediatric CT imaging. This scoping review emphasizes advanced RD reduction methods used in pediatric CT head imaging for diagnosing various clinical conditions with optimum RD and diagnostic image quality (IQ). A detailed search was conducted across five databases, such as PubMed, Scopus, CINAHL, Web of Science (WOS) and Embase using relevant keywords. A total of 24 articles were included in the final review. RD parameters and IQ related data were extracted from each article. Conventional RD reduction techniques in CT such as reducing the tube voltage, tube current and other scanning parameters, face limitations particularly in the pediatric population. These techniques lead to a trade-off between a lower RD and poor IQ which might obfuscate diagnostic details due to decreased contrast resolution with greater image noise and artifacts. To balance RD and diagnostic IQ, advanced technologies such as iterative reconstruction (IR) and deep learning image reconstruction (DLIR) algorithms with ultra-low dose protocols are increasingly being used. Hence, the review concludes that, compared with conventional dose reduction techniques, artificial intelligence based DLIR algorithms enhance IQ even for ultra-low dose protocols across various clinical domains in pediatric CT imaging.