Ordered-subsets Expectation-maximization Reconstruction Algorithm Research Articles

Improved [68Ga]Ga-PSMA-11 image qualities reconstructed by total variation regularized expectation maximization on total-body PET/CT.

Total variation regularized expectation maximization (TVREM) reconstruction algorithm on the image quality of gallium (68GA) prostate-specific membrane antigen-11 ([68Ga]Ga-PSMA-11) total-body positron emission tomography/computed tomography (PET/CT). Images of a phantom with small hot sphere inserts and the total-body PET/CT scans of 51 prostate cancer patients undergoing [68Ga]Ga-PSMA-11 were reconstructed using TVREM with 5 different penalization factors between 0.09 and 0.45 and for 20-, 40-, 60-, 120-, and 300-second acquisition, respectively. As a comparison, the same data were also reconstructed using the ordered subset expectation maximization (OSEM) with 3 iterations, 20 subsets, and 300 second acquisition. The contrast recovery coefficients (CRC) and background variability (BV) of the phantom, the tumor-to-background ratios (TBR), the contrast recovery (CR) ratio, the image noise of the liver, and maximum standard uptake value (SUVmax) of the lesions were calculated to evaluate the image quality. The clinical performance of the images was evaluated by 2 radiologists with a 5-point scale (1-poor, 5-excellent). The TVREM reconstructions groups fwith 120 second acquisition and the penalization of 0.27 to 0.45 showed the best performance in terms of CR, TBR, image noise, and the gain of SUVmax compared to that obtained in the OSEM 300 second group. Even the image noise of the TVREM 120 second group with a penalization factor of 0.27 and 0.36 was comparable to the OSEM 300 second group; the lesions' SUVmax increased by 28% whereas the image noise decreased by 5% and 14%, respectively. The TVREM 120 second group with a penalization factor of 0.36 (5.00±0.00) had the highest qualitative score that equaled OSEM and TVREM for the 300 second (P>0.05) group. Our study has shown the potential of the TVREM reconstruction algorithm with optimized penalization factors to achieve comparable [68Ga]Ga-PSMA-11 total-body PET/CT image quality with a shorter acquisition time, compared with the conventional OSEM reconstruction algorithm.

Parallax error in long-axial field-of-view PET scanners—a simulation study

There is a growing interest in the design and construction of a PET scanner with a very long axial extent. One critical design challenge is the impact of the long axial extent on the scanner spatial resolution properties. In this work, we characterize the effect of parallax error in PET system designs having an axial field-of-view (FOV) of 198 cm (total-body PET scanner) using fully-3D Monte Carlo simulations. Two different scintillation materials were studied: LSO and LaBr3. The crystal size in both cases was 4 × 4 × 20 mm3. Several different depth-of-interaction (DOI) encoding techniques were investigated to characterize the improvement in spatial resolution when using a DOI capable detector. To measure spatial resolution we simulated point sources in a warm background in the center of the imaging FOV, where the effects of axial parallax are largest, and at several positions radially offset from the center. Using a line-of-response based ordered-subset expectation maximization reconstruction algorithm we found that the axial resolution in an LSO scanner degrades from 4.8 mm to 5.7 mm (full width at half max) at the center of the imaging FOV when extending the axial acceptance angle (α) from ±12° (corresponding to an axial FOV of 18 cm) to the maximum of ±67°—a similar result was obtained with LaBr3, in which the axial resolution degraded from 5.3 mm to 6.1 mm. For comparison we also measured the degradation due to radial parallax error in the transverse imaging FOV; the transverse resolution, averaging radial and tangential directions, of an LSO scanner was degraded from 4.9 mm to 7.7 mm, for a measurement at the center of the scanner compared to a measurement with a radial offset of 23 cm. Simulations of a DOI detector design improved the spatial resolution in all dimensions. The axial resolution in the LSO-based scanner, with α = ± 67°, was improved from 5.7 mm to 5.0 mm by incorporating a two-layer DOI detector. These results characterize the maximum axial blurring for a fully open 2 m long PET scanner and demonstrate that large sensitivity gains are possible with a modest reduction in resolution when using current clinical detector technology with no DOI capability.

Evaluation of simultaneous 201Tl/99mTc dual-isotope cardiac SPECT imaging with model-based crosstalk compensation using canine studies

BackgroundSimultaneous 201Tl/99mTc-sestamibi dual-isotope myocardial perfusion SPECT imaging can reduce imaging time and produce perfectly registered rest/stress images. However, crosstalk from 99mTc into 201Tl images can significantly reduce 201Tl image quality. We have developed a model-based compensation (MBC) method to compensate for this crosstalk. The method has previously been validated with phantom and simulation studies. In this study, we evaluated the MBC method using a canine model. MethodsLeft anterior descending or left circumflex coronary artery stenoses were created in 50 adult mongrel dogs weighing 20-30 kg. The dogs were injected with 111 MBq (3 mCi) of 201Tl at rest, and a SPECT study acquired. Stress was induced by administering adenosine to the dog, followed by injection of 740 MBq (20 mCi) of 99mTc-sestamibi at peak stress. A second SPECT study was performed with data acquired in both 201Tl and 99mTc energy windows to provide simultaneous dual-isotope projection data. The images were reconstructed using the ordered-subsets expectation-maximization reconstruction algorithm with compensation for attenuation, scatter, and detector response. For simultaneously acquired 201Tl data, we also applied the MBC method to compensate for crosstalk contamination from 99mTc. ResultsWithout compensation, 99mTc crosstalk increased the estimated 201Tl activity concentration in the rest images and reduced defect contrast. After MBC, the 201Tl images were in good agreement with the registered single-isotope images and ex vivo count data. The ischemic (IS) to non-ischemic (NIS) region 201Tl activity concentration ratios were computed for single-isotope and dual-isotope studies. The correlation with ex vivo IS-NIS ratios was 0.815 after MBC, compared to the 0.495 from data without compensation. In addition, the regression line for the IS-NIS ratios with MBC was almost parallel to the line of identity with a slope of 0.93, compared to a slope of 0.45 without compensation. ConclusionsThese results demonstrate that model-based crosstalk compensation can provide substantial reduction of crosstalk effects in simultaneously acquired myocardial perfusion SPECT images in living biological systems.

Performance measurement of PSF modeling reconstruction (True X) on Siemens Biograph TruePoint TrueV PET/CT

The Siemens Biograph TruePoint TrueV (B-TPTV) positron emission tomography (PET) scanner performs 3D PET reconstruction using a system matrix with point spread function (PSF) modeling (called the True X reconstruction). PET resolution was dramatically improved with the True X method. In this study, we assessed the spatial resolution and image quality on a B-TPTV PET scanner. In addition, we assessed the feasibility of animal imaging with a B-TPTV PET and compared it with a microPET R4 scanner. Spatial resolution was measured at center and at 8 cm offset from the center in transverse plane with warm background activity. True X, ordered subset expectation maximization (OSEM) without PSF modeling, and filtered back-projection (FBP) reconstruction methods were used. Percent contrast (% contrast) and percent background variability (% BV) were assessed according to NEMA NU2-2007. The recovery coefficient (RC), non-uniformity, spill-over ratio (SOR), and PET imaging of the Micro Deluxe Phantom were assessed to compare image quality of B-TPTV PET with that of the microPET R4. When True X reconstruction was used, spatial resolution was <3.65 mm with warm background activity. % contrast and % BV with True X reconstruction were higher than those with the OSEM reconstruction algorithm without PSF modeling. In addition, the RC with True X reconstruction was higher than that with the FBP method and the OSEM without PSF modeling method on the microPET R4. The non-uniformity with True X reconstruction was higher than that with FBP and OSEM without PSF modeling on microPET R4. SOR with True X reconstruction was better than that with FBP or OSEM without PSF modeling on the microPET R4. This study assessed the performance of the True X reconstruction. Spatial resolution with True X reconstruction was improved by 45 % and its % contrast was significantly improved compared to those with the conventional OSEM without PSF modeling reconstruction algorithm. The noise level was higher than that with the other reconstruction algorithm. Therefore, True X reconstruction should be used with caution when quantifying PET data.

High quality ordered subset expectation maximization reconstruction algorithm based on multi-resolution for PET images

In Positron Emission Tomography(PET) imaging,Maximum Likelihood Expectation Maximization(MLEM) algorithm cannot be directly applied to clinical diagnosis due to suppressing noise ineffectively and converging slowly.Although Ordered Subset Expectation Maximization(OSEM) algorithm converges fast,it will lead to a significant decline in the quality of the reconstructed image.To address this problem,multi-resolution technology was introduced into the subset of the OSEM reconstruction algorithm to suppress noise and stabilize solving process.The experimental results indicate that the new algorithm overcomes the drawback of the traditional algorithm on degrading the reconstructed image and has the advantage of fast convergence.The proposed reconstruction algorithm can obtain a higher Signal-to-Noise Ratio(SNR) and a superior visual effect.

High Resolution ${}^{125}$I Pinhole SPECT Imaging of the Mouse Thyroid With the MediSPECT Small Animal CdTe Scanner

The first in vivo tomographic 125I imaging of the mouse thyroid carried out with the new MediSPECT small animal SPECT scanner is presented. The scanner is based on a fine pitch CdTe semiconductor pixel detector (14 × 14 mm2, 256 × 256 square pixel with a 55 m side) and equipped with a set of high resolution collimators. The collimation and detection units of the scanner are mounted on a gantry, rotating around a horizontal axis, along which is placed the small animal housing. In an in vivo test, the mouse was injected with a Na125 I solution having a total activity of 31.8 MBq. The planar projections for SPECT reconstruction were acquired with a 300 m pinhole (magnification 1.47 and field of view of 9.6 × 9.6 mm2). The projections were captured in a step-and-shoot fashion and were processed with an Ordered Subsets-Expectation Maximization reconstruction algorithm in order to obtain the SPECT images. Several 125I imaging tests have been made by using phantoms to assess the detector spatial resolution. The measured spatial resolution with a 300 m pinhole is about 0.5 mm in planar imaging and better than 1 mm in tomographic imaging.

Simultaneous assessment of cardiac perfusion and function using 5-dimensional imaging with Tc-99m teboroxime

Dynamic single photon emission computed tomography (SPECT) acquisition and reconstruction of early poststress technetium 99m teboroxime washout images has been shown to be useful in the detection of coronary disease. Assessment of poststress regional wall motion may offer additional use in assessing coronary disease. Our goal was to investigate the feasibility of simultaneously imaging myocardial ischemia and transient poststress akinesis using gated-dynamic SPECT. A gated-dynamic mathematical cardiac torso (MCAT) phantom was developed to model both teboroxime kinetics and cardiac regional wall motion. A lesion was simulated as having delayed poststress teboroxime washout together with a transient poststress wall motion abnormality. Gated projection data were created to represent a 3-headed SPECT system undergoing a total rotation of 480 degrees . The dynamic expectation-maximization reconstruction algorithm with postsmoothing across gating intervals by Wiener filtering, and the ordered-subset expectation maximization reconstruction algorithm with 3-point smoothing across gating intervals were compared. Compared with the ordered-subset expectation maximization with 3-point smoothing, the dynamic expectation-maximization algorithm with Wiener filtering was able to produce visually higher-quality images and more accurate left ventricular ejection fraction estimates. From simulations, we conclude that changing cardiac function and tracer localization possibly can be assessed by using a gated-dynamic acquisition protocol combined with a 5-dimensional reconstruction strategy.

A channelized-hotelling-trace collimator design method based on reconstruction rather than projections

In this study a new SPECT (single photon emission computed tomography) collimator design technique is investigated. In this feasibility study, the collimator hole size is the only design variable changed. Other collimator parameters are fixed to a low-energy-high-resolution (LEHR) design. The design is based on a task of imaging a small hot lesion with a uniform background. A channelized hotelling trace (CHT) is used to quantify the lesion detectability in a reconstructed image for a particular collimator parameter. The unique feature of this development is that the image quality is not evaluated for planar projection data, but rather is evaluated for the reconstructed tomographic image with an OS-EM (ordered-subset expectation-maximization) reconstruction algorithm. Our results suggest that a collimator hole size that is larger than the LEHR collimator hole size is preferable for lesion detection. This implies that the parameter that gives the optimal lesion detectability on the planar projections may not give the optimal lesion detectability in the reconstructed image. In SPECT, the optimal collimator design should be reconstruction algorithm dependent.

Rotating slant-hole single-photon emission-computed tomography

This dissertation presents a detailed investigation of the use of rotating slant-hole (RSH) collimators in single-photon emission-computed tomography(SPECT). We propose RSH SPECT as an alternative imaging technique, which may have advantages over more traditional SPECT techniques, in certain imaging applications. The basic imaging properties of RSH collimators are derived. This includes the derivation of expressions for the size and location of the common volume of view and the geometric resolution and efficiency. In addition, we derive an analytical expression for the geometric response and transfer functions of RSH collimators. The theoretical expressions for the aforementioned imaging characteristics are verified via Monte Carlo simulations and phantom experiments. Next, we present the development of a three-dimensional (3-D) analytic reconstruction algorithm for RSH SPECT. This includes the derivation of both a backprojection of the filtered projections and a filtering of the simple backprojections (BpjF) algorithm. The relationship between the derived 3-D RSH reconstruction filter and a number of other reconstruction filters is discussed. In addition, the implementation of the BpjF reconstruction algorithm is described and its performance is demonstrated via simulations and phantom experiments. We have also implemented an iterative ordered-subsets expectation-maximization (OS-EM) reconstruction algorithm for RSH SPECT. A detailed description of the development of the projector/backprojector employed by the OS-EM algorithm is provided. Simulation and phantom experiments are again used to demonstrate the performance of the projector/backprojector and the iterative OS-EM reconstruction algorithm. Rotating slant-hole SPECT has potential applications in cardiac and breast imaging. Thus, we demonstrate the feasibility, advantages, and disadvantages of using RSH SPECT via a comparison with conventional SPECT in a cardiac and breast imaging phantom experiment. In conclusion, we summarize the results presented and discuss their implications on the current and future state of RSH SPECT research.