Encrypted login | home

Program Information

Alternative Analytic Solution to the Paralyzable Detector Model to Calculate Deadtime and Deadtime Loss


W Siman

W Siman, S Kappadath , UT MD Anderson Cancer Center, Houston, TX

Presentations

SU-C-9A-4 Sunday 1:00PM - 1:55PM Room: 9A

Purpose:
Some common methods to solve for deadtime are (1) dual-source method, which assumes two equal activities; (2) model fitting, which requires multiple acquisitions as source decays; and (3) lossless model, which assumes no deadtime loss at low count rates. We propose a new analytic alternative solution to calculate deadtime for paralyzable gamma camera.

Methods:
Deadtime T can be calculated analytically from two distinct observed count rates M1 and M2 when the ratio of the true count rates alpha=N2/N1 is known. Alpha can be measured as a ratio of two measured activities using dose calibrators or via radioactive decay. Knowledge of alpha creates a system with 2 equations and 2 unknowns, i.e., T and N1. To verify the validity of the proposed method, projections of a non-uniform phantom (4GBq 99mTc) were acquired in using Siemens SymbiaS multiple times over 48 hours. Each projection has >100kcts. The deadtime for each projection was calculated by fitting the data to a paralyzable model and also by using the proposed 2-acquisition method. The two estimates of deadtime were compared using the Bland-Altmann method. In addition, the dependency of uncertainty in T on uncertainty in alpha was investigated for several imaging conditions.

Results:
The results strongly suggest that the 2-acquisition method is equivalent to the fitting method.
The Bland-Altman analysis yielded mean difference in deadtime estimate of ~0.076us (95%CI: -0.049us, 0.103us) between the 2-acquisition and model fitting methods. The 95% limits of agreement were calculated to be -0.104 to 0.256us. The uncertainty in deadtime calculated using the proposed method is highly dependent on the uncertainty in the ratio alpha.

Conclusion:
The 2-acquisition method was found to be equivalent to the parameter fitting method. The proposed method offers a simpler and more practical way to analytically solve for a paralyzable detector deadtime, especially during physics testing.


Contact Email: