Radiology and Oncology

Background: Scintimammography (SM) is a promising functional radionuclide imaging technique that is generally undertaken using high resolution parallel-hole collimators with Gamma Cameras. The main clinical limitation of this technique is inaccuracy in detecting small lesions less than 1 cm diameter. This limitation is due to resolution-efficiency trade-off that is inherent in the use of collimation. As an alternative approach this study proposes using a simple Coded Aperture (CA) mask, instead of a collimator, coupled to a standard clinical gamma camera for breast tumor imaging. This imaging technique successfully predicts the overall form of artefacts arising from the near-field imaging geometries.

 

Aim & Methods: To investigate the applications of CA technique a Monte Carlo Simulation (MCS) is used using MCNPX package. To emulate SM, 3D pseudo-anthropomorphic phantoms have been developed and verified and used along with a realistic model of a clinical gamma camera. This study examines a moderately compressed breast phantom in a cranio caudalprojection.

 

The performance of such an imaging system is modeled by the MCS method and images are reconstructed by correlation analysis. This imaging system was quantitatively evaluated using variable parameters: The detected photon from tumor, spatial resolution, photon statistics and lesion visibility of the system at several tumor-background activity ratios. The effectiveness and the performance of the CA-SM system was assessed and compared with low energy high resolution parallelhole collimator and ultra-high resolution parallel-hole collimator image formation systems.

 

Results: The predicted background can be used to correct the near-field effect of 3D sources, as might be found in SM using CA. The simulated planar images from these collimator-based image formation systems suggest tumors of 1 cm diameter may be observable with a tumor-background-ratio of 5:1. However, when the tumor diameter is ≤0.8 cm these become less reliable detecting small (less than 1 cm in diameter) lesion unless a tumor-background-ratio of more than 10:1 is used.

 

Conclusion: The results of the simulations demonstrate that with near-field artefacts corrections the CA-SM approach shows

good performance in lesion detection for all lesions (located 3 cm deep in a 6 cm thick breast phantom) and for a tumorbackground ratio as low as 3:1. This level of performance is highly competitive, in some cases, superior to conventional

collimator based image formation methods.