This study aimed to assess radiation dose reduction efficacy and image quality impact of automatic tube current modulation (ATCM) combined with deep-learning reconstruction algorithm “ClearInfinity” in low-dose pediatric non-contrast head CT scans.

This study retrospectively analyzed imaging data from 160 pediatric patients who underwent non-contrast head CT scans. The patients were divided into two groups according to the scanning protocol: group A (n=80) used the low-dose reconstruction protocol based on the ATCM technology combined with the ClearInfinity algorithm; Group B (n=80) adopted the conventional-dose reconstruction protocol using fixed parameters combined with iterative reconstruction. The patients were divided into four age groups, with 20 cases in each group. Radiation dose was evaluated using metrics such as volume CT dose index (CTDI_vol), dose-length product (DLP), and effective dose (ED) to compare the radiation exposure levels between the two groups. Image quality assessment included both objective and subjective evaluations. For the objective evaluation, regions of interest (ROIs) with an area of approximately 20 mm² were selected at four anatomical levels to calculate the signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) for each region. Subjective evaluation was performed by two experienced physicians using a blinded method to rate the overall image quality, the degree of anatomical detail display, and the control of posterior cranial fossa artifacts.. Statistical analysis was conducted using SPSS software.

The result shows that there was no statistically significant difference in the baseline data between the two groups. Radiation dose metrics, including CTDI_vol, DLP, and ED, as well as objective image evaluation indicators such as SNR and CNR in different regions, all showed statistically significant differences (P<0.05). The radiation doses in Group A were all significantly lower than those in Group B. The subjective evaluation results indicated that the overall image quality of group A was improved compared to group B (P<0.001), the posterior cranial fossa artifacts were optimized (P<0.001), and the consistency of the evaluations by the two physicians was excellent (ICC=0.891~0.923, P<0.001).

The ATCM technology combined with the ClearInfinity algorithm can significantly reduce the radiation dose of pediatric head CT scans. Although the objective evaluation indicates a slight decline in image quality for older age groups, the subjective evaluation shows an improvement in image quality, achieving a coordinated optimization of dose and image quality, and is expected to provide a new paradigm for low-dose CT scans of the pediatric head.