Optimizing Dose Management: Implementing and Evaluating Auto-Exposure Control Systems

Thesis Abstract

The optimization of dose management in radiography hinges on addressing the variability in patient exposure while maintaining diagnostic image quality, with Auto-Exposure Control (AEC) systems representing a pivotal technological approach to standardize dose delivery across diverse patient cohorts and imaging tasks. This study aims to optimize dose management by implementing and evaluating AEC systems within clinical radiography workflows, with specific objectives to (1) quantify dose reductions achieved through standardized AEC settings across varying body regions, (2) assess the impact of AEC on image quality metrics and diagnostic confidence, (3) evaluate operator interaction with AEC interfaces and its influence on exposure outcomes, (4) develop a practical decision-support framework for selecting AEC parameters tailored to patient size and clinical indication, and (5) model the relationship between AEC performance, patient dose, and image quality using a comprehensive analytical approach. The investigation is anchored in the Health Belief Model and the Technology Acceptance Model to interpret user adoption and behavioral responses to AEC deployment, complemented by lauded radiographic dose optimization principles such as ALARA (As Low As Reasonably Achievable). A mixed-methods design is employed, combining a quasi-experimental trial with a longitudinal observational component conducted across five radiography departments in three tertiary hospitals. The population comprises adult patients undergoing standard chest, abdomen, and extremity radiographs. A total sample of 1,200 imaging examinations (approximately 240 per body region) will be collected over six months, with purposive sampling of imaging protocols to represent diverse patient sizes (kVp, mAs ranges) and clinical indications. Quantitative data will be drawn from DICOM metadata, including dose-area product (DAP), entrance surface dose (ESD), and exposure indices, alongside objective image quality scores obtained via standardized tolerances and modulation metrics. AEC parameter configurations, including detector configuration, back-up timer, and sensitivity thresholds, will be logged to correlate with dose and image quality outcomes. Qualitative data will be collected through semi-structured interviews with radiographers (n=30) and radiology residents (n=15) to elucidate interface usability, workflow impact, and perceived reliability of AEC decisions. Quantitative analysis will proceed with descriptive statistics to characterize baseline dose and image quality, followed by multilevel linear mixed-effects models to evaluate the effect of AEC on DAP/ESD while controlling for patient size, body region, and technique factors. ANOVA will compare image quality scores across AEC settings, with post-hoc contrasts to identify optimal parameter ranges. Regression analysis will model the relationship between AEC performance indicators (consistency of exposure, adaptation to patient anatomy) and diagnostic confidence as rated by radiologists. Time-series analysis will examine dose trends over the six-month period to detect learning effects and interface adaptations. The qualitative data will be analyzed using thematic analysis, coded against the Technology Acceptance Model constructs and workflow integration themes, with triangulation to relate user perceptions to quantitative outcomes. Ethical approval will be obtained, with informed consent from radiographers participating in interviews and strict de-identification of patient data. Expected findings include measurable reductions in mean DAP and ESD without compromising objective image quality scores or diagnostic confidence, improved consistency of exposure across patient cohorts, and enhanced radiographer satisfaction linked to intuitive AEC interfaces. The study anticipates identifying critical AEC parameter ranges that balance dose minimization with image fidelity across chest, abdomen, and extremity protocols, and revealing workflow factors that influence successful AEC adoption. The contribution to knowledge lies in providing empirical evidence on the dose-saving potential and practical implementation considerations of AEC systems within real-world radiography settings, bridging gaps between theoretical dose optimization models and clinical practice, and offering a validated decision-support framework for parameter selection customized to patient size and clinical indication. The main conclusion is that thoughtfully configured AEC systems can substantially reduce patient radiation exposure while preserving diagnostic image quality, provided that radiographers receive targeted training, interface design aligns with clinical workflows, and ongoing monitoring sustains parameter calibration. Recommendations include standardizing AEC configuration guidelines across departments, integrating decision-support prompts within radiography information systems, implementing routine auditing of AEC performance, and extending the study to pediatric cohorts and tomosynthesis contexts to generalize findings across radiographic modalities.