The Effect Of Post-Reconstruction Gaussian Filter On Image Quality Of Iterative Reconstruction In SPECT/CT

One of the medical imaging modalities used in nuclear medicine is Single Photon Emission Computed Tomography/Computed Tomography (SPECT/CT). SPECT/CT can produce gamma ray distribution images and show the location of radionuclides in the patient's body. SPECT/CT images are obtained from a reconstruction process, one of which is the iterative reconstruction method. Iterative reconstruction is divided into two algorithms, namely the Maximum Likelihood Expectation Maximization (MLEM) algorithm and the Ordered-Subsets Expectation Maximization (OSEM) algorithm. This study aims to evaluate the addition of a post-reconstruction Gaussian filter on the image quality of SPECT/CT reconstruction and to compare the image quality produced from low and high iterations in reconstruction. This study is a retrospective descriptive study using 20 thyroid scintigraphy patients using the SPECT/CT modality. Sinogram data from 20 patients will be reconstructed using the MLEM and OSEM algorithms with low iterations (4 iterations) and high iterations (30 iterations) and the addition of a post-reconstruction Gaussian filter. Image quality was evaluated based on the calculation of the Contrast to Noise Ratio (CNR) and Signal to Noise Ratio (SNR). The results showed that image quality with low iterations was higher than that with high iterations in iterative reconstruction without filters. The addition of a post-reconstruction Gaussian filter can improve the image quality of iterative reconstruction. Image quality in the MLEM algorithm is more optimal than in the OSEM algorithm, but it takes longer than the OSEM algorithm. Therefore, it can be concluded that the OSEM algorithm with low iterations and the addition of a post-reconstruction Gaussian filter can be used in clinical examinations.