A Phantom Study and a Retrospective Clinical Analysis to Investigate the Impact of a New Image Processing Technology
Jaydev K Dave1, David J Eschelman1, Jenna R Wasserman2
1Department of Radiology, Thomas Jefferson University, Main Building, 132 S. 10th St., Suite 1092B, Philadelphia, PA, 19107, Lancaster, Pennsylvania.
Purpose:
To investigate changes in radiation dose and image quality using phantoms and hepatic embolization procedures performed with a new image processing technology (ClarityIQ) for a single-plane flat-detector-based interventional fluoroscopy system.
Materials And Methods:
Phantom study was performed using acrylic sheets simulating different patient sizes. Air kerma rates (AKRs) were compared for different fluoroscopy modes and magnification modes without and with ClarityIQ. Repeat hepatic embolization procedures performed on the same lobe of the liver in the same patient by the same interventional radiologist between January 2013 and July 2014 without and with ClarityIQ were evaluated retrospectively. This included treatment of 33 hepatic lobes in 26 patients. Cumulative air kerma (CAK), kerma-area product (KAP), and factors affecting radiation dose were extracted from study metadata and compared. Blinded randomized image quality review was performed on arteriograms using a five-point scale.
Results:
The phantom study revealed a significantly lower AKR (P < .005) with ClarityIQ. Repeated-measures analysis revealed a significant effect of ClarityIQ (P ≤ .001) on CAK and KAP, with reductions ranging between 9% and 85% (median, 67%) and between 5% and 89% (median, 75%), respectively, on a case-by-case basis. Mean reductions in CAK and KAP were 279 mGy and 134,030 mGy·cm(2), respectively. Image quality review scores were significantly lower (P ≤ .001) with ClarityIQ, effecting visualization of tumor vasculature and appearance of noise texture.
Conclusions:
ClarityIQ resulted in radiation dose reduction in the phantom study and in the hepatic embolization procedures, but with a decrease in subjective perceptions of image quality.
