Advanced AI Methods for Dose Reduction in Interventional and Diagnostic Fluoroscopy

Purpose: To provide an educational overview of how emerging artificial intelligence (AI) technologies contribute to radiation dose optimization across CT and fluoroscopic imaging workflows.

Methods/Materials: A broad review of current literature was conducted to identify key applications of AI related to radiation dose reduction in CT and fluoroscopy. Themes examined included AI‑assisted patient positioning, acquisition optimization, automated region‑of‑interest selection, equipment performance monitoring, and AI‑based reconstruction methods.

Results: AI contributes to dose reduction through several mechanisms:

- Improved patient positioning and scan planning: AI can reduce unnecessary repeat scans by guiding correct patient alignment and optimizing CT acquisition parameters.

- Targeted fluoroscopy exposure: AI-enabled fluoroscopy systems can automatically select and expose only the pathology-containing region, lowering dose to adjacent tissues and reducing overall exposure.

- Optimization of imaging workflows: AI-driven decision support tools can help refine imaging protocols and standardize practices, reducing operator‑dependent variability.

- Quality assurance and equipment monitoring: Many dose-related challenges stem from outdated imaging equipment and the absence of structured QA programs; AI can help identify performance degradation early and assist in maintaining optimal system functionality.

- Integration into reconstruction and post-processing: Current research more broadly shows that deep learning-based CT reconstruction can significantly reduce dose while preserving diagnostic quality, supporting safer protocols across both CT and fluoroscopy.

Conclusions: AI represents a meaningful advancement in radiation protection strategies for both CT and fluoroscopy. By improving scan planning, patient positioning, ROI targeting, and equipment quality assurance, AI addresses several modifiable contributors to excess radiation exposure. As these technologies continue to evolve, AI is poised to become an essential component of modern imaging safety and education.