Radiation Dose Optimization in CT Scans: A Machine Learning Perspective

Computed tomography (CT) is a widely used diagnostic imaging technique, but concerns over excessive radiation exposure have driven the need for dose optimization. Despite advancements, existing methods struggle to balance image quality with radiation dose reduction. This study addresses this gap by employing machine learning models to predict and optimize radiation dose levels based on empirical dosimetry data. Five machine learning algorithms, including logistic regression and ensemble learning techniques, were trained on patient-specific imaging parameters to estimate optimal dose levels. Findings indicate that machine learning models significantly enhance dose prediction accuracy while maintaining diagnostic image quality. The results highlight the potential of AI-driven strategies to reduce radiation risks in CT imaging and improve clinical decision-making. This research underscores the necessity of integrating intelligent dose optimization systems into routine radiology practice for safer and more efficient patient care.