model based, anatomy dependent method for ultra-fast creation of primary SPECT projections

Monte Carlo [MC] is the most common method for simulating virtual SPECT projections. It is useful for optimizing procedures, evaluating correction algorithms and more recently image reconstruction as a forward projector in iterative algorithms; however, the main drawback of MC is its long run time. We introduced a model based method considering the effect of body attenuation and imaging system response for fast creation of noise free SPECT projections. Collimator detector response [CDR] was modeled by layer by layer blurring of activity phantom using suitable Gaussian functions. Using the attenuation phantom, in each angle, attenuation factor [AF] was calculated for each voxel. This calculated AF is the weight for the emission voxel and states the detection probability of photons that are emitted from that voxel. Finally weighted ray sum of the blurred phantom was driven to create a projection. For the next projection, our phantom was rotated and the procedure was repeated until all projections were acquired. Root Mean Square error [RMS] between all 60 modelled projection and real MC simulated projections was decreased from 0.58 +/- 0.15 using simple Radon to 0.19 +/- 0.03 using our suggested model. This value was 0.56 +/- 0.16 using blurred Radon without attenuation modelling, and 0.21 +/- 0.03 using attenuated Radon without CDR modelling. Our suggested model that considers the effect of both attenuation and CDR simultaneously results in more accurate analytical projections compared with conventional Radon model. Creation of 60 primary SPECT projections in less than one minute may make this method as a proper alternative for MC simulation. This model can be used as a forward projector during iterative image reconstruction for correction of CDR and attenuation that is necessary for quantitative SPECT

Assessment of the imact of applying attenuation correction on the accuracy of activity recovery in Tc99m-ECD brain SPECT of healthy subject using statistical parameters mapping [SPM]

Photon attenuation in tissues is the primary physical degrading factor limiting both visual qualitative interpretation and quantitative analysis capabilities of reconstructed Single Photon Emission Computed Tomography [SPECT] images. The aim or present study was to investigate the effect of attenuation correction on the detection of activation foci following statistical analysis was SPM. The study population consisted of twenty normal subjects [11 male, 9 female, and age 30-40 years]. SPECT images were reconstructed using filter back projection and attenuation correction was done by the Chang method. The SPECT imagings was obtained 20 min after intravenous injection of 740-1110 MBq [20-30 mCi] of Tc99m-ECD and were acquired on 128x128 matrices with a 20% symmetric energy window at 140 keV. These data publicly distributed by the Society of Nuclear Medicine of Toronto Hospital. the data was standardized with respect to the Montreal Neurological institute [MNI] atlas with a 12 parameter affine transformations. Images were then smoothed by a Gaussian filter of 10 mm FWHM. Significance differences between SPECT images were estimated at every voxel using statistical t-test and p-value as the significant criteria was set at 0.05. The contrast comparing non attenuation corrected images suggest that regional brain perfusion activity increase in the cerebrum, frontal [T-value 12.06], temporal [T-value 10.63] and occipital [T-value 9.31] lobe and decrease in the sub-lobar, extra-nuclear [T-value 17.46] and limbic lobe, posterior cingulated [T-value 17.46] before attenuation correction compare with attenuation correction. It can be concluded that applying correction in brain SPECT can effectively improve the accuracy of the detection of activation are [p<0.05]

[Compensation of cross-contamination in simultaneous 201Tl/99mTc myocardial perfusion SPECT imaging]

It is a common protocol to use 201Tl for the rest and 99mTc for the stress cardiac SPECT imaging. Theoretically, both types of imaging may be performed simultaneously using different energy windows for each radionuclide. However a potential limitation is the cross-contamination of scattered photons from 99mTc and collimator X-rays into the 201Tl energy window. We used a middle energy window method to correct this cross-contamination. Using NCAT, a typical software torso phantom was generated. An extremely thin line source of 99mTc activity was placed inside the cardiac region of the phantom and no activity in the other parts. The SimSET Monte Carlo simulator was used to image the phantom in different energy windows. To find the relationship between projections in different energy windows, deconvolution theory was used. We investigated the ability of the suggested functions in three steps: Monte Carlo simulation, phantom experiment and clinical study. In the last step, SPECT images of eleven patients who had angiographic data were acquired indifferent energy windows. All of these images were compared by determining the contrast between a defect or left ventricle cavity and the myocardium. We found a new 2D kernel which had an exponential pattern with a much higher center. This function was used for modeling 99mTc down scatter distribution from the middle window image. X-ray distribution in the 201Tl window was also modeled as the 99mTc photopeak image convolved by a Gaussian function. Significant improvements in the contrast of the simultaneous dual 201Tl images were found in each step before and after reconstruction. In comparison with other similar methods, better results were acquired using our suggested functions. Our results showed contrast improvemtn in thallium images after correction, however, many other parameters should be evaluated for clinical approaches. There are many advantages in simultaneous dual isotope imaging. It halves imaging time and reduces patient waiting time and discomfort. Identical rest/stress registration of images also facilitates physicists' motion or attenuation corrections and physicians' image interpretation

[Energy window setting for optimum Tl-201 cardiac imaging]

Poor sensitivity and poor signal to noise ratio because of low injected thallium dose and presence of scattered photons are the main problems in using thallium in scintigraphic imaging of the heart. Scattered photons are the main cause of degrading the contrast and resolution in SPECT imaging that result in error in quantification. Thallium decay is very complicated and photons are emitted in a wide range of energies of 68-82 keV. It seems possible to achieve better primary to scattered radiation ratio and better image sensitivity simultaneously if the energy window setting is carefully selected. This investigation was performed in three steps: Monte Carlo simulation, phantom experiment and clinical study. In simulation step, the new 4D digital NCAT phantom was used to simulate the distribution of activity [201Tl] in patient torso organs. The same phantom was used to simulate the attenuation coefficient of different organs of the typical patient's body. Two small defects on different parts of left ventricle also were generated for further quantitative and qualitative analysis. The simulations were performed using the SimSET simulator to generate images of such patient. The emissions arising from Tl-201 decay were simulated in four steps using the energies and relative abundances. Energy spectra for primary and scatter photons were calculated. Changing the center and width of energy windows, optimum energy window characteristics were determined. In next step jaszczak phantom was prepared and used for SPECT imaging in different energy windows. In last step SPECT images of 7 patients who had angiographic data were acquired in different energy windows. All of these images were compared qualitatively by four nuclear medicine physicians independently. The optimum energy window was determined as a wider asymmetric window [77keV?30%] that its center is not placed on photo-peak of energy spectrum. This window increased the primary counts rate and PTSR considerably as compared with the conventional symmetric energy window [67keV%]. In a comparison which performed between clinical images acquired in suggested 77-30% window with conventional 67-20% window, a considerable increase was found in myocardial to defect contrast [1.541 +/- 0.368] and myocardial to cavity contrast [1.171 +/- 0.099]. A negligible increase was also found in total counts of images using this window. We found that conventional symmetric energy window [67keV +/- 10%] couldn't be a suitable choice for thallium heart imaging; furthermore three energy windows, 73keV-30%, 75keV-30% and 77keV-30%, were determined as optimum window options. For further analysis the images from such windows were compared in each three steps of this investigation. In all steps conventional symmetric energy window [67keV-20%] was introduced as the worst case and the asymmetric 77keV-30% was determined as the most suitable

Reducing the respiratory motion artifacts in pet cardiology: a simulation study

There are several technical features that make PET an ideal device for the noninvasive evaluation of cardiac physiology. Organ motion due to respiration is a major challenge in diagnostic imaging, especially in cardiac PET imaging. These motions reduce image quality by spreading the radiotracer activity over an increased volume, distorting apparent lesion size and shape and reducing both signal and signal-to-noise ratio levels 4D average male torso [2 cm diaphragmatic motion] produced by NCAT phantom was used for simulations. Emission sinograms generated by Eidolon PET simulator were reconstructed using iterative algorithm using STIR. The respiratory motion correction [RMC] applied to data sets using an automatic algorithm. Cross section views, activity profiles, contrast-to-noise ratios and left ventricle myocardium widths of corrected and non-corrected images were compared to investigate the effect of applied correction. Comparison of respiratory motion corrected and non corrected images showed that the algorithm properly restores the left ventricle myocardium width, activity profile and improves contrast-to-noise ratios in all cases. Comparing the contrast recovery coefficient []shows that the applied correction effected phases of number 7,8 and 9 of cardiac cycle more than the other 13 phases and the maximum value being 1.43 +/- 0.07 for phase number 8. The maximum value of ratio of the left ventricle myocardium width for non-RMC and RMC images along the line profile passing the apicobasal direction and along the line profile passing from the middle of the lateral wall of the left ventricle were 1.38 +/- 0.07 for phase number 9 and 1.12 +/- 0.03 for phases of number 8 and 9 respectively. Blurring and ghosting of each image depends on the speed of diaphragm during that respiratory phase. This simulation study demonstrates that respiratory motion correction has good overall effect on PET cardiac images and can reduces errors originating from diaphragmatic motion and deformation. Effect of such a correction varies from one cardiac phase to another and this depends on the blurring and ghosting of all respiratory phases used to form this cardiac phase. Using an automatic algorithm capable of correcting respiratory motion using full signal may be very useful to prevent lengthening of the overall scan time to obtain same motionless lesion signal levels