Quantification of PET and CT misalignment errors due to bulk motion in cardiac PET/CT imaging: phantom and clinical studies

Non simultaneous acquisition between CT and PET module can introduce misalignment artefact in cardiac PET/CT imaging due to patient motion. We assessed the clinical impact of patient motion and the resulting mismatch between CT and corresponding CT-based attenuation corrected [CTAC] PET images on apparent myocardial uptake values in cardiac PET/CT imaging. The evaluation of patient motion was performed using clinical and experimental phantom studies acquired on the Biograph TP 64 PET/CT scanner. In order to simulate patient motion, CT images were manually shifted from 0 to 20 mm in steps of 5-mm in six different directions. The reconstructed PET images using shifted CT were compared with the original PET images. The assessment of PET images was performed through qualitative interpretation by an experienced nuclear medicine physician and through quantitative analysis using volume of interest based analysis. Moreover, Box and Whisker plots were calculated and bull's eye view analysis performed. PET images were also reoriented along the short, horizontal and vertical long axis views for a better qualitative interpretation. A 20-mm shift in the right direction between attenuation and PET emission scans produced mean absolute percentage difference in uptake values in the lateroanterior [33.42 +/- 9.07] and lateroinferior [27.39 +/- 10.43] segments of the myocardium. Misalignment could introduce artifactual nonuniformities in apparent myocardial uptake value and the variations were more significant for the misalignment toward the right, feet and head directions, in such a way that even with a 5-mm shift in the CT image, errors in interpretation of PET images could occur. Furthermore, errors in PET uptake estimates were observed for movements as large as 10-mm in the left, posterior and anterior directions

comparative assessment of dynamic and conventional thallium-201 SPECT myocardial perfusion imaging: monte carlo simulations and case studies

Purpose: Clinical myocardial perfusion SPECT is commonly performed using static imaging. Dynamic SPECT enables extraction of quantitative as well as relative perfusion information. We aimed to evaluate the ability of dynamic SPECT for regular perfusion assessment in comparison to conventional SPECT in the context of thallium-201. Methods: Simulations were performed utilizing a 4D-NCAT phantom for a dual-head gamma camera via the SIMIND Monte-Carlo simulator. 64s acquisition time-frames were used to track these dynamic changes. Different summations of time-frames were performed to create each dataset, which were compared to a standard static dataset. In addition, the effect of different delay-times post-injection was assessed. Twenty-segment analysis of perfusion was performed via the QPS analyser. Dynamic data were subsequently acquired in clinical studiesusing simulation-optimized protocols. Results: For different summations of time-frames, perfusion scores in the basal and mid regions revealed 14.4% and 7.3% increases in dynamic SPECT compared to conventional imaging, with maximum changes in the basal anterior, while the distal and apical segments did not show noticeable changes. Specifically, dynamic imaging including 4 to 6 time-frames yielded enhanced correlation [R=0.957] with conventional imaging, in comparision to the usage of less time frames. Greatest correlation with conventional imaging was obtained for post-injection delays of 320 to 448s [R=0.982 to R=0.988]. Conclusion: While dynamic SPECT opens up an important opportunity for quantitative assessment [e.g. via generation of kinetic parameters], it was shown to generate highly consistent perfusion information compared to established conventional imaging. Future work focuses on merging these two important capabilities

Direct 4D Parametric Imaging in Dynamic Myocardial Perfusion PET

Purpose: Dynamic myocardial perfusion [MP] PET imaging followed by tracer kinetic modeling allows quantification of myocardial blood flow, thus enabling computation of the coronary flow reserve, with considerable clinical potentials. Nonetheless, utilization of short dynamic frames can lead to noisy flow estimates, an issue that is further amplified in parametric imaging at the voxel level. Our purpose is to utilize an enhanced image reconstruction framework to better address this issue. Methods: We implemented a novel 4D reconstruction scheme to directly estimate MP parametric images from the measured dynamic datasets. This included formulation of a 4D log-likelihood objective function relating the kinetic parameters to the projection datasets, and implementing numerical methods to optimize the objective function. We also utilized the technique of optimization transfer to enable more convenient and reliable parametric imaging. We simulated MP Rb-82 PET projection datasets based on the XCAT phantom utilizing patient-based time activity curves for the various organs and clinically realistic noise levels, followed by noise vs. bias analysis. Results: The proposed direct 4D methodology was shown to outperform conventional indirect parametric imaging, reducing noise by over 50% [matched bias], with further reductions of 15% in noise and a factor of five speed-up when optimization transfer was additionally utilized. Conclusion: Direct 4D PET image reconstruction is a viable and very promising approach towards robust parametric MP PET imaging at the individual voxel level.

Monte-Carlo based characterization of the counting rate (NECR) response for rersonalized optimization of the administered activity in clinical PET imaging

Purpose: The statistical quality of a PET scan can be significantly affected by the associated patient and scanner characteristics. Standard protocols could be optimized by regulating the administered activity Aadm such that the statistical quality is maximized for each individual patient for a given scan time. The objective is to model the direct relationship between the noise equivalent count rate NECR and Aadm for a wide range of scanner and patient parameters employed in clinical scans. Methods: A series of extensive and validated Monte Carlo simulations is utilized to systematically investigate, under realistic and controlled conditions, the effect of a wide set of [i] phantom sizes modeling children, slim and obese patients, [ii] bed positions, [iii] energy windows, [iv] coincidence time windows, [v] and combination of dead times and detector responses on the NECR for a range of Aadm. Results: A wide plateau is observed in NECR[Aadm] curves particularly for large patients, suggesting that 90-95% of peak NECR can still be obtained with considerably less Aadm. Moreover, for default scanner configurations and cardiac beds, an optimal Aadm range of 55-65 MBq for HR+ and 300-450 MBq for Biograph scanners, with the maximum NECR being considerably higher for the latter.Conclusions: The generalized NECR[Aadm] model can be utilized to predict for each individual Optimization, patient scan an optimal range of Aadm for which NECR is maximized, thus potentially allowing [a] for efficient utilization of the available activity in PET centers and [b] for minimization of cumulative radiation exposure

Performance comparison of four commercial GE discovery PET/CT scanners: a monte carlo study using GATE

Combined PET/CT scanners now play a major role in medicine for in vivo imaging in oncology, cardiology, neurology, and psychiatry. As the performance of a scanner depends not only on the scintillating material but also on the scanner design, with regards to the advent of newer scanners, there is a need to optimize acquisition protocols as well as to compare scanner performances on an objective basis. In this study we evaluate and compare the performance of 4 Commercial GE PET/CT cameras, the [i] BGO-based Discovery ES PET/CT [DLS], [ii] the BGO-based Discovery ST PET/CT [DST], [iii] the BGO-based Discovery STE PET/CT [DSTE] and finally [iv] the EYSO-based Discovery RX PET/CT [DRX] scanner using the Geant4 Application for Tomographic Emission [GATE]. GATE is an open source Monte Carlo simulation platform developed for PET and SPECT studies and is supported by the OpenGATE collaboration. In accordance with the National Electrical Manufactures Association [NEMA] NU 2-2001 protocols, the validation of models is carried out against actual published measurements and the performance comparison is done for sensitivity, scatter fraction and count rate performance, showing very similar performance compared with published results, thus enabling investigations to better model system performance [e.g. resolution degradation] within the reconstruction task.. The simulated results demonstrate highest sensitivity performance with the DST [though with the highest scatter fraction], and highest NECR performance for the EYSO-based DRX, The results also show that DRX, DES and DSTE PET/CT cameras have nearly the same amount of scatter fraction

4D PET: beyond conventional dynamic PET imaging

In this paper, we review novel techniques in the emerging field of spatiotemporal 4D PET imaging. We will discuss existing limitations in conventional dynamic PET imaging which involves independent reconstruction of dynamic PET datasets. Various approaches that seek to attempt some or all of these limitations are reviewed in this work, including techniques that utilize iterative temporal smoothing, advanced temporal basis functions, principal components transformation of the dynamic data, wavelet-based techniques as well as direct kinetic parameter estimation methods. Extension of 4D PET to 5D PET in which the additional dimension of [respiratory or cardiac] gating is considered has also been discussed

PET vs. SPECT: in the context of ongoing developments [review article]

This paper intends to compare the abilities of the two major imaging modalities in nuclear medicine imaging: Positron Emission Tomography [PET] and Single Photon Emission Computed Tomography [SPECT]. The motivations are many-fold: [i] To gain a better understanding of the strengths and limitations of the two imaging modalities in the context of recent and ongoing developments in hardware and software design; [ii] To emphasize that certain issues, historically and commonly thought as limitations, may now be instead viewed as challenges that can be addressed; [iii] To point out that existing PET and SPECT scanners in the field can [much] benefit from improvements in image-reconstruction software; [iii] To point-out [to engineers, physicists and software-developers] important areas of research in PET and SPECT imaging that will be instrumental to further improvements in the two modalities

overview of clinical PET/CT

This article is intended to provide an overview of various aspects of clinical PET/CT. These include discussions of: [i] Important areas of clinical application; [ii] Opportunities in clinical research; [iii] Scanner and operating-mode considerations [e.g. BGO vs. LSO, LYSO or GSO scanners, 2D vs. 3D imaging]. [iv] Study-specific considerations [e.g. patient preparation and positioning issues, injected dose, use of CT contrast agents]

Advanced motion correction methods in PET [review article]

With the arrival of increasingly higher resolution PET systems, small amounts of motion can cause significant blurring in the images, compared to the intrinsic resolutions of the scanners. In this work, we have reviewed advanced correction methods for the three cases of [i] unwanted patient motion, as well as motions due to [ii] cardiac and [iii] respiratory cycles. For the first type of motion [most often studies in PET brain imaging], conventional motion-correction algorithms have relied on extraction of the motion information from the emission data itself. However, the accuracy of motion compensation in this approach is degraded by the noisy nature of the emission data. Subsequently, advanced methods, as reviewed in this work, make use of external real-time measurements of motion. Various image-based and projection-based correction methods have been discussed and compared. The paper also reviews recent and novel applications that perform corrections for cardiac and respiratory motions. Unlike conventional gating schemes, in which the cardiac and respiratory gated frames are independently reconstructed [resulting in noisy images], the reviewed methods are seen to follow a common trend of seeking to produce images of higher quality by making collective use of all the gated frames [and the estimated motion]. As an observation, a general theme in motion-correction methods is seen to be the use of increasingly sophisticated software to make use of existing advanced hardware. In this sense, this field is very open to future novel ideas [hardware, and especially software] aimed at improving motion detection, characterization and compensation