Program Information
Consistent Low Contrast Detectability and Optimal Dose for Different Patient Sizes in Abdominal CT
Y Zhou*, J Nute , A Scott , C Lee , Cedars-Sinai Medical Center, Los Angeles, CA
Presentations
SU-G-206-9 (Sunday, July 31, 2016) 4:00 PM - 6:00 PM Room: 206
Purpose: To study optimal dose dictated by consistent low contrast detectability for various patient sizes in abdominal CT.
Methods: Seven realistically shaped tissue-equivalent abdomen phantoms (CIRS TE01–TE08) were utilized. Helical scans were performed on each phantom with GE 750 HD and Siemens mCT 64 at different dose levels. The images were reconstructed using comparable kernels (standard and Bf35 for GE and Siemens, respectively). The noise images, obtained using subtraction from adjacent slices, were partitioned into matrices of square cells matching different sizes of lesions (1.8 – 10 mm). The noise granularities were quantified by the standard deviation (σ) of the matrix cell pixel mean values distribution at each cell size. The minimum detectable contrast (MDC), defined as 3.29σ with 95% confidence of detectability, was obtained for each phantom at different doses. We proposed that the MDC is inversely proportional to the lesion size and to the square root of the size specific dose and the proportional factor is a function of phantom size. The MDC data was fitted against the lesion size and dose following the proposed model.
Results: The simplified power law model was found to fit the minimum detectable contrast against the lesion and dose very well for various phantoms (R^2 > 0.983). The proportionality factor was found to fit the phantom equivalent diameter (L) to a Gaussian amplifier with R2 = 0.983 and to an exponential function with R^2 = 0.999, for GE HD750 and Siemens mCT 64, respectively. The parameters of the Gaussian amplifier varied with different scanners. To obtain the tube current-rotation time product (mAs) from the size specific dose (D), the relationship of D per 100 mAs versus L at 120 kVp was also presented for each scanner.
Conclusion:The relationship of low contrast detectability and optimal dose across different patient sizes was successfully obtained.
Contact Email: