osem algorithm

Nowadays, imaging of the blood supply of the heart muscle by single photon emission computed tomography (SPECT: Single Photon Emission Computed Tomography) due to its non-invasive nature and providing information with physiological value and low cost compared to the valuable angiography method. It is highly diagnostic. But these images undergo changes and artifacts under the influence of factors, the result of which is the reduction of the diagnostic accuracy of the images and false positive cases. During the detection process, several physical effects such as attenuation, scattering, and collimator response function affect the frequency of emitted photons; this leads to the destruction of the contrast and as a result of reducing the quantitative and qualitative accuracy of the images. Attenuation, as the most destructive factor of SPECT images, reduces the quality of SPECT images of heart blood supply and reduces the sensitivity of tests related to the diagnosis of coronary artery diseases, and for non-uniform environments, especially in nuclear imaging of chest areas. And the heart is necessary to produce a map of patient attenuation coefficients. The existence of scattered photons is also one of the main factors of error in quantization; the detection of scattered events affects the contrast of the lesions and causes the lack of image resolution and signal-to-noise ratio. Therefore, to correct the attenuation and scattering of the rays in the heart images quantitatively and qualitatively, patterns are needed in SPECT systems. Due to the importance of the topic, various research groups around the world have presented their research and results on correcting the effect of scattering of rays and also correcting the effect of weakening the rays. If there was no limitation of energy resolution, it was easily possible to identify the scattered rays and prevent them from being recorded in the image. Because we know that scattered rays lose energy. Because gamma rays are single energy and their energy amount is completely known. Therefore, each photon with less energy will represent scattered rays, but due to the limited energy resolution of the gamma camera, a range is usually considered on the sides of the main energy, which is called the energy window. It is assumed that the photons recorded in this energy range are primary photons, but in fact, many photons scattered in the body are also recorded in this window. These scattered rays do not carry correct spatial information and lead to a decrease in image resolution and contrast and quantization errors in the image. In nuclear medicine, instead of researching and examining the patient or processing the image of the patient, simulated images can be examined. Simulators can provide information about each of the image destruction factors. The purpose of this research is to propose a new method for scattering correction, in this research, a combination of Monte Carlo and modeling is used for the rapid production of scattered views, and in the proposed method, the two-matrix method is used, this method At the stage of generating mathematical views, dispersion is added and this problem leads to the removal of scattered rays. As a result, an image is reconstructed that is free from the effects of attenuation and non-ideal dispersion and leads to an increase in contrast and improvement of power. Detecting waste, increasing the signal-to-noise ratio, and increasing the accuracy of quantification.

In this study, the effect of applying attenuation and dispersion correction using two energy windows (DEW) and three energy windows (TEW) methods in cardiac aspect imaging was investigated and evaluated, and to simulate cardiac aspect imaging, a special code similar to SAR Monte Carlo GATE was used as the SPECT imaging system and XCAT digital phantom with activity distribution and realistic attenuation map was used to model the human trunk.

Comparison of image contrast improvement in different modes of attenuation and dispersion correction shows that the highest image contrast is obtained from the (TEW1+AC) method with an average increase of 25% and MSE in different modes of attenuation correction. And the dispersion compared to the reference image was reduced from 51.5% to 54.5%. Compared to the reference image, MSE decreased from 1.4 in Un_Cor to 1.15, 1.13, 1.12, and 1.14 in AC+TEW1, AC+DEW, AC, and AC+TEW2, respectively, and the signal-to-noise ratio (SNR) increased up to 71% in all methods of applying dispersion correction along with attenuation correction compared to applying attenuation correction (AC).

In this study, the effect of attenuation and dispersion correction in 5 non-correction modes, with attenuation correction, attenuation, and dispersion correction using two-window and three-window methods with triangular approximation and three-window with trapezoidal approximation on We evaluated XCAT phantom simulated images and heart muscle perfusion images by SPECT method and 4 different parameters were used to compare and evaluate the images, including profile, contrast, mean squared error (MSE) and signal to noise. According to the results of the quantification of reconstructed images, it is possible to apply dispersion correction along with attenuation correction.