A comparative study of scatter effects in ¹²⁵I and ^99mTc quantitative pinhole SPECT

Abstract

1677

Objectives: The goal of this study is to rebut the misconception that scatter effects are negligible for quantitative ¹²⁵I small animal pinhole SPECT.

Methods: A comparative Monte Carlo simulation study was performed using GATE. We simulated pinhole projection data of mouse and rat size cylinders filled with water with either a ^99mTc or ¹²⁵I point source at the center. A modular camera with pixellated NaI(Tl) crystals and a pinhole collimator was simulated for data acquisition. The energy resolutions of the detector were 12.5% and 25.3%, and the energy windows were 20% and 15-45 keV for ^99mTc and ¹²⁵I, respectively. The effects of attenuation and scatter were quantified by comparing with source-in-air simulations.

Results: As expected, attenuation losses for primary photons were greater for ¹²⁵I than for ^99mTc (e.g., -43% vs. -18% for mouse). However, contrary to previous claims, the scatter-to-primary ratio (SPR) of the detected photons within the energy window was noticeably larger for ¹²⁵I (38.4%, vs. 24.8% in ^99mTc). Further analysis indicated that, for ¹²⁵I, the backscatter edge (~24 kev) is very close to the main photopeak (27 kev) resulting in the large amount of scatter detected in the energy window. Without compensating for attenuation and scatter, errors in activity estimates for the mouse phantom were -24.1% (¹²⁵I) and -12% (^99mTc), respectively.

Conclusions: Achieving accurate quantitation for ¹²⁵I requires attenuation compensation and measures to reduce the effects of scatter. Due to the energy resolution and compact nature of the scatter energy spectrum, narrowing of energy windows is not a viable way to reduce the effects of scatter. It is thus important to further characterize the scatter response functions for the purpose of implementing scatter compensation techniques.

Research Support: NIH EB1558