Variability in radiation dose and image quality: A comparison across fluoroscopy-system vendors, generations of equipment and institutions.

OBJECTIVES: To evaluate differences in radiation dose and image quality across institutions, fluoroscope vendors and generations of fluoroscopes for pediatric cardiac catheterization. BACKGROUND: Increased recognition of the potentially harmful effects of ionizing radiation has spurred technological advances in fluoroscopes, as well as increased focus on optimizing fluoroscope performance. There is currently little understanding of variability in the dose-image quality relationship across institutions, fluoroscope vendor and/or generation of equipment. METHODS: We evaluated latest generation fluoroscopes from Phillips, Siemens, GE, and Toshiba, and an older generation Phillips fluoroscope (release date 2003) at three different institutions. Radiation dose was measured using an anthropomorphic dose-assessment phantom with effective dose in mSv estimated from Monte Carlo simulations. Image quality phantom images were scored on a 12-point scale by three blinded reviewers. RESULTS: Fluoroscope effective doses ranged from 0.04 to 0.14 mSv/1,000 pulses for fluoroscopy with associated composite image quality scores ranging from 8.0 ± 0.6 to 10.4 ± 1.3. For cineangiography, effective doses ranged from 0.17 to 0.57 mSv/1,000 frames with image quality scores ranging from 10.1 ± 0.3 to 11.1 ± 0.3. There was modest correlation between effective dose and image quality (r = 0.67, P = 0.006). The older generation fluoroscope delivered consistently higher doses than the newer generation systems (2.3- to 3.5-fold higher for fluoroscopy; 1.1- to 3.4-fold higher for cineangiography) without appreciable differences in image quality. CONCLUSION: Technological advances have markedly improved fluoroscope performance. Comparing latest generation systems across vendors and institutions, we found variability in the dose-IQ relationship and speculate that this reflects both equipment and institutional optimization practices.

Three dimensional dose distribution comparison of simple and complex acquisition trajectories in dedicated breast CT.

PURPOSE: A novel breast CT system capable of arbitrary 3D trajectories has been developed to address cone beam sampling insufficiency as well as to image further into the patient's chest wall. The purpose of this study was to characterize any trajectory-related differences in 3D x-ray dose distribution in a pendant target when imaged with different orbits. METHODS: Two acquisition trajectories were evaluated: circular azimuthal (no-tilt) and sinusoidal (saddle) orbit with ±15° tilts around a pendant breast, using Monte Carlo simulations as well as physical measurements. Simulations were performed with tungsten (W) filtration of a W-anode source; the simulated source flux was normalized to the measured exposure of a W-anode source. A water-filled cylindrical phantom was divided into 1 cm(3) voxels, and the cumulative energy deposited was tracked in each voxel. Energy deposited per voxel was converted to dose, yielding the 3D distributed dose volumes. Additionally, three cylindrical phantoms of different diameters (10, 12.5, and 15 cm) and an anthropomorphic breast phantom, initially filled with water (mimicking pure fibroglandular tissue) and then with a 75% methanol-25% water mixture (mimicking 50-50 fibroglandular-adipose tissues), were used to simulate the pendant breast geometry and scanned on the physical system. Ionization chamber calibrated radiochromic film was used to determine the dose delivered in a 2D plane through the center of the volume for a fully 3D CT scan using the different orbits. RESULTS: Measured experimental results for the same exposure indicated that the mean dose measured throughout the central slice for different diameters ranged from 3.93 to 5.28 mGy, with the lowest average dose measured on the largest cylinder with water mimicking a homogeneously fibroglandular breast. These results align well with the cylinder phantom Monte Carlo studies which also showed a marginal difference in dose delivered by a saddle trajectory in the central slice. Regardless of phantom material or filled fluid density, dose delivered by the saddle scan was negligibly different than the simple circular, no-tilt scans. The average dose measured in the breast phantom was marginally higher for saddle than the circular no tilt scan at 3.82 and 3.87 mGy, respectively. CONCLUSIONS: Not only does nontraditional 3D-trajectory CT scanning yield more complete sampling of the breast volume but also has comparable dose deposition throughout the breast and anterior chest volume, as verified by Monte Carlo simulation and physical measurements.