On the Relationship of Minimum Detectable Contrast to Dose and Lesion Size in Abdominal CT
Y Zhou*, A Scott, J Allahverdian, Cedars-Sinai Medical Center, Los Angeles, CA 90048TU-C-103-3 Tuesday 10:30AM - 12:30PM Room: 103
Purpose: CT dose quantification is typically guided by pixel noise or contrast-to-noise ratio that does not delineate low contrast details well. To optimize the dose for lesion specific diagnosis, we utilize the statistically defined minimum detectable contrast (MDC) to study it's relationship to dose and lesion size in typical abdomen CT scans.
Methods:An abdomen phantom (CIRS TE-07) simulating the medium sized patient was used. The phantom contains a cylindrical void of 4 cm diameter, filled with the low contrast insert containing six cylindrical targets (1% and 2% contrast) of 1.2 mm to 7 mm. Helical scans were conducted using a Siemens Sensation 64 (mCT) with 120 kVp and mAs ranged 35- 595 (kernel B30s, pitch 1.4 and width 5 mm). The uniform sections of the contiguous slices were subtracted and the resulted regions were divided by grids of cells sized from 1.37 mm- 6.84 mm. Standard deviations (sd) were computed from the means of all cells and scaled down by square root of 2. The MDC, quantified by the signal to background difference equal to 3.29sd, was fitted to dose D (phantom specific CTDIvol) and cell sizes (d). For validation, independent observations on the targets were made and compared with the measured contrast. MDC versus noise and lesion size was also identified.
Results: MDC versus D and d was fitted to a power law (R square 0.993), with power indices of -0.537 and -1.027, respectively. The findings were validated by the independent observations on the larger targets whose contrast can be measured. The MDC was found to be proportional to noise and inversely proportional to d (R square 0.994).
Conclusion:The relationships of MDC versus dose and lesion size, and MDC versus noise and lesion size were established. They can be used for task specific dose optimization.
Add this talk to vcal | ical | Contact Email: