Click here to


Are you sure ?

Yes, do it No, cancel

Linear Regression for Modeling Common Fluoroscopy Dose Metrics in Cardiac Catheterization Procedures

S Mann*, Z Wegermann, M Russ, T Wei, WS Jones, Duke University Medical Center, Durham, NC,


(Sunday, 7/12/2020)   [Eastern Time (GMT-4)]

Room: AAPM ePoster Library

Purpose: Fluoroscopy is an integral part of cardiac catheterization procedures; however, there exists the potential for variable modality utilization that depends on operator tendencies and patient characteristics. Additionally, modern fluoroscopy systems employ advanced image processing that may help to reduce radiation dose. This study aims to assess the impact of operators, patient BMI, and fluoroscopic technology on common fluoroscopy dose metrics within a single institution.

Methods: Diagnostic left- and right-heart catheterization data over a 9 month period were retrospectively collected from 5 Philips Allura fluoroscopy systems. Systems differed in detector size (10” and 20”), configuration (4 single-plane units, 1 bi-plane unit) and image processing technology (1 Clarity vs 4 non-Clarity). Utilization data were retrieved via Radiation Dose Structured Reports. Collected procedure data included attending and assisting physicians, patient BMI, and procedure type. Metrics of utilization were extracted including air kerma (AK), dose area product (DAP), and fluoroscopy time (FT), among others. Data were binned according to procedure and log-transformed before analysis using linear regression models that included the following parameters: Fluoroscopy System, Attending & Assisting Physician, and Patient BMI.

Results: R-squared values ranged from 0.14-0.33, with the lower values observed for FT models. For all procedures, the Clarity-enabled system demonstrated a reduction in AK (20-49%) and DAP (26-60%) despite similar performance in FT. Performance across operators revealed modest variability. Patient BMI was significant for AK and DAP, with a compounding increase of 0.3-1.5% for each increasing point of BMI.

Conclusion: Fluoroscopic system technology plays the largest role in reducing radiation dose while having minimal impact on fluoroscopy usage as measured by FT. For the investigated diagnostic cardiac catheterization procedures, minimal differences were observed among attending and assisting physicians. Further work is needed to investigate radiations dose and safety practices in interventional cardiovascular procedures, including the impact of physician experience.


Fluoroscopy, Dose, Linear Regression Analysis


IM- X-Ray: Fluoroscopy, digital angiography, and DSA

Contact Email