Room: Exhibit Hall | Forum 1
Purpose: To evaluate a daily image quality control (QC) regimen in a busy academic interventional radiology (IR) department using signal-to-noise (SNR) ratio from fluoroscopic loops.
Methods: Daily QC was performed for a month on six Siemens fluoroscopes (two Artis Zeego, two Axiom Artis and two Artis Q) using a 10â€?x10â€?x3â€? custom-built patient equivalent phantom consisting of polycarbonate, copper and aluminum. The phantom was placed on the table in the same position each day and centered under fluoro, with a source-to-image distance set to 100 cm, and with the table raised to just meet the receptor. The programs in each room were set to the default â€œNormalâ€? 7.5 pulse per second fluoro. After actuating fluoro several times to allow the filter selection to be reproducible, a fluoro loop was acquired for 5 seconds and stored using the â€œStore Fluoroâ€? function. Fluoro loops were sent to a QC-Track (Atirix Medical Systems) server for automated processing. A 12 mm region of interest (ROI) was placed in a uniform region in the center of the phantom. SNR was computed in the ROI for each frame of the last 2 seconds of the fluoro loop. The SNRâ€™s for each of the last 15 frames were then averaged to yield a single SNR for the loop.
Results: Using statistical process control logic, QC for all units and times fell within +/- 3 standard deviations of the mean for that unit. The SNR is stable for 3 different vendor models over 6 different IR suites over a month long period.
Conclusion: Automated and observer-independent quality control of units used during fluoroscopically guided interventions was piloted for a busy IR department. Minimal technologist effort and change in workflow were needed to regularly monitor system performance and readiness for the day.