Dose optimization and modeled lifetime attributable cancer risk in brain CT using hybrid statistical iterative reconstruction

Background Radiation dose in brain computed tomography (CT) should be monitored. The low-dose computed tomography (LDCT) protocol is an employed strategy for reduction in patients’ effective dose ( E ). Methods Data were collected from four emergency units employing different CT scanners: Siemens, Philips, General Electric (GE), and Canon. Each scanner utilizes hybrid statistical iterative reconstruction (H/SIR) algorithm combined with LDCT protocols (LDCT-H/SIR), whereas the standard-dose protocol utilizes filtered back projection (FBP) algorithm (SDCT-FBP). A total of 152 (75 Males, 77 Females) patients were scanned using SDCT-FBP, whereas 142 (69 Males, 73 Females) patients were scanned using the LDCT-H/SIR. A validated software tool for CT dosimetry (IndoseCT) automatically extracts volume computed tomography dose index (CTDI vol ) and dose length products (DLP) in order to calculate size-specific dose estimation (SSDE). Corrected DLP (DLPc) and E are used for cancer risk estimation based on the BEIR VII report. Results Effective dose ( E ) and LAR decreased significantly for integrated LDCT with H/SIR (LDCT-H/SIR) compared to SDCT-FBP ( p < 0.001). The mean percentage reduction in E across all vendors was 48.3% (40.85% for Philips, 44.44% for Siemens, 57.08% for GE, and 61.20% for Canon). Based on BEIR VII models, the model-projected LAR was lower for LDCT-H/SIR, corresponding to projected relative reduction of 44% for males and 58% for females. The estimated attributable risk difference (ARD) was 4.0 cases per 100, 000 for males and 8.0 cases per 100, 000 for females. Qualitative image quality scores remained comparable between protocols ( p > 0.05). Conclusion Protocol optimization with integrated LDCT-H/SIR yields substantial radiation savings without compromising image quality across vendors. These findings reflect the theoretical, model-projected reductions in radiation-attributable cancer risk rather than observed clinical incidence, highlighting the population-level benefit of dose optimization.