Document Type: Original Paper
Ph.D. Student in Medical Physics, Tarbiat Modarres University, Tehran, Iran.
Assistant Professor, Medical Physics Dept., Tarbiat Modarres University, Tehran, Iran.
M.D., Dept. of Nuclear Medicine, Rajaei Heart Centere, Tehran, Iran.
Associate Professor, Dept. of Biostatistics, Tarbiat Modarres University, Tehran, Iran.
Introduction: The purpose of this study is to define the optimal parameters for the tomographic
reconstruction procedure in a routine single photon emission tomography. The Hoffman brain phantom
is modified to evaluate the reconstruction method. The phantom was imaged in a 3 and 2-dimensional
conformation and the results were compared.
Materials and Methods: The 2D phantom (slices) was imaged independently. The tomographic
image was performed using the phantom in 3D conformation. Planar image was used as the reference
image for each corresponding reconstructed slice image. A set of projection data with different count
densities were obtained from 99mTc with the phantom. The data were transformed into interfile format
and transferred to personal computer. A software was designed in Matlab (Version 7) to support the
phantom and the automatic calculation. All the data were reconstructed using back-projection
technique with the ramp, Shepp-Logan, Cosine, Hamming, Hanning (each filter having so different
parameters), Butterworth, Metz and Wiener (each filter having 2000 different parameters). The quality
of the images was estimated by comparing the reconstructed images with the reference image by the
use of universal image quality index (UIQI).
Results: All the filters tested score substantially better than the control (P<0.05). The data from this
study suggests that the Hanning and Hamming are the weakest of the seven filter types tested (mean of
UIQI= 0.2218, 0.2219). This is due to response of Hanning and Hamming filters at the low
frequencies. The quantitative results of this study indicate that Wiener filter provided best image
quality of the phantom than the other filters regardless of which collimator is used. The mean of UIQI
was the most for wiener filter in comparison with the other filters (P<0.05).
Discussion and Conclusion: These results clearly show that Wiener filter improves quality image and
the extent of the improvement has been quantified. The optimum parameters obtained for the filters
show direct relation with total counts in the study.