Individualized Radiation Dose Assessment in Low-Dose Chest CT: The Role of DLPss and Topogram Optimization

Background: The increasing use of computed tomography (CT) has led to a substantial rise in population exposure to ionizing radiation, highlighting the need for accurate and individualized dose assessment methods. This study aimed to evaluate a novel dosimetric parameter-the size-specific dose-length product (DLPss)-in low-dose chest CT (LDCT) protocols and to compare its performance with conventional dose metrics. Methods: A retrospective single-center analysis was conducted in a cohort of 221 patients undergoing LDCT of the chest. Anthropometric parameters were used to calculate the size-specific conversion factor (k), enabling determination of SSDE and DLPss. Dose parameters (CTDIvol, DLP, SSDE, and DLPss) were analyzed and compared with data from a standard chest CT cohort (n = 134) from the first study in the series. The contribution of the topogram to total radiation dose was also assessed. All examinations were considered diagnostically adequate in routine clinical evaluations. Results: The mean CTDIvol in the LDCT group was 1.33 mGy, with a DLPss of 61.93 mGy·cm and an estimated effective dose below 0.7 mSv, representing a dose reduction exceeding 82% compared to standard CT. DLPss values were approximately 23% higher than conventional DLP, indicating underestimation of dose by standard metrics. The topogram accounted for 10.23% of total radiation dose in LDCT, significantly higher than in standard CT (1.84%). Significant sex-related differences were observed in CTDIvol, DLP, and DLPss, but not in SSDE. Conclusions: DLPss provides a more comprehensive and individualized assessment of radiation exposure than conventional dose metrics by integrating patient size and scan length. The substantial contribution of the topogram to total dose in LDCT highlights the need for its optimization, particularly in long-term screening programs. From a clinical perspective, implementation of DLPss may improve patient-specific risk stratification and support more precise monitoring of cumulative radiation exposure, especially in populations undergoing repeated imaging, such as lung cancer screening cohorts. Advanced reconstruction algorithms, including deep learning-based methods, may enable further dose reductions and warrant future clinical investigation.