Yunlong Zan

Fast direct estimation of the blood input function and myocardial time activity curve from dynamic SPECT projections via reduction in spatial and temporal dimensions

PURPOSE Reconstruction of parametric images from dynamic single photon emission computed tomography (SPECT) data acquired with slow rotating cameras is a challenge because the estimation of the time-activity curves (TACs) may involve fitting data to an inconsistent underdetermined system of equations. This work presents a novel algorithm for the estimation of the blood input function and myocardial TAC with high accuracy and high efficiency directly from these projections. METHODS In the proposed dynamic reconstruction method, the information from the segmentation of functional regions from the static reconstructed image was used as a prior to construct a sparse matrix, through which the spatial distribution of the radioactive tracer was represented. Then the temporal distribution of the radioactive tracer was modeled by nonuniform B-spline basis functions which were determined according to a new selection rule. With reduction in both the spatial and temporal dimensions of the reconstructed image, the blood input function and myocardial TAC were estimated using the 4D maximum likelihood expectation maximization algorithm. The method was validated using data from both digital phantom simulations and an experimental rat study. RESULTS Compared with the conventional dynamic SPECT reconstruction method without the reduction in spatial dimensions, the proposed method provides more accurate TACs with less computation time in both phantom simulation studies and a rat experimental study. CONCLUSIONS The proposed method is promising in both providing more accurate time-activity curves and reducing the computation time, which makes it practical for small animal studies using clinical systems with slow rotating cameras.

Longitudinal Evaluation of Sympathetic Nervous System and Perfusion in Normal and Spontaneously Hypertensive Rat Hearts with Dynamic Single-Photon Emission Computed Tomography.

The objective of this work was to evaluate the sympathetic nervous system and structure remodeling during the progression of heart failure in a rodent model using dynamic cardiac single-photon emission computed tomography (SPECT). The spontaneously hypertensive rat (SHR) model was used to study changes in the nervous system innervation and perfusion in the left ventricular (LV) myocardium with the progression of left ventricular hypertrophy (LVH) to heart failure. Longitudinal dynamic SPECT studies were performed with seven SHR and seven Wistar-Kyoto (WKY) rats over 1.5 years using a dual-head SPECT scanner with pinhole collimators. Time-activity curves (TACs) of the 123I-MIBG and 201Tl distribution in the LV blood pool and myocardium were extracted from dynamic SPECT data and fitted to compartment models to determine the influx rate, washout rate, and distribution volume (DV) of 123I-MIBG and 201Tl in the LV myocardium. The standardized uptake values (SUVs) of 123I-MIBG and 201Tl in the LV myocardium were also calculated from the static reconstructed images. The influx and washout rates of 123I-MIBG did not show a significant difference between SHRs and WKY rats. The DVs of 123I-MIBG were greater in the SHRs than in the WKY rats (p = .0028). Specifically, the DV of 123I-MIBG became greater in the SHRs by 6 months of age (p = .0017) and was still significant at the age of 22 months. The SUV of 123I-MIBG in SHRs exhibited abnormal values compared to WKY rats from the age of 18 months. There was no difference in the influx rate and the washout rate of 201Tl between the SHRs and WKY rats. The SHRs exhibited greater DV of 201Tl than WKY rats after the age of 18 months (p = .034). The SUV of 201Tl in SHRs did not show any significant difference from WKY at all ages. The higher DV of 123I-MIBG in the LV myocardium reveals abnormal nervous system activity of the SHRs at an age of 6 months, whereas a greater DV of 201Tl in the LV myocardium can only be detected at an age of 18 months. The results show that the abnormal nervous system activity appears earlier than perfusion. Furthermore, the comparison between the DV and the SUV indicates that dynamic SPECT with 123I-MIBG and 201Tl with the kinetic parameter DV is capable of detecting abnormalities of the LV at an early age.

Dynamic FDG-PET Imaging to Differentiate Malignancies from Inflammation in Subcutaneous and In Situ Mouse Model for Non-Small Cell Lung Carcinoma (NSCLC)

Background [18F]fluoro-2-deoxy-D-glucose positron emission tomography (FDG-PET) has been widely used in oncologic procedures such as tumor diagnosis and staging. However, false-positive rates have been high, unacceptable and mainly caused by inflammatory lesions. Misinterpretations take place especially when non-subcutaneous inflammations appear at the tumor site, for instance in the lung. The aim of the current study is to evaluate the use of dynamic PET imaging procedure to differentiate in situ and subcutaneous non-small cell lung carcinoma (NSCLC) from inflammation, and estimate the kinetics of inflammations in various locations. Methods Dynamic FDG-PET was performed on 33 female mice inoculated with tumor and/or inflammation subcutaneously or inside the lung. Standardized Uptake Values (SUVs) from static imaging (SUVmax) as well as values of influx rate constant (Ki) of compartmental modeling from dynamic imaging were obtained. Static and kinetic data from different lesions (tumor and inflammations) or different locations (subcutaneous, in situ and spontaneous group) were compared. Results Values of SUVmax showed significant difference in subcutaneous tumor and inflammation (p<0.01), and in inflammations from different locations (p<0.005). However, SUVmax showed no statistical difference between in situ tumor and inflammation (p = 1.0) and among tumors from different locations (subcutaneous and in situ, p = 0.91). Values of Ki calculated from compartmental modeling showed significant difference between tumor and inflammation both subcutaneously (p<0.005) and orthotopically (p<0.01). Ki showed also location specific values for inflammations (subcutaneous, in situ and spontaneous, p<0.015). However, Ki of tumors from different locations (subcutaneous and in situ) showed no significant difference (p = 0.46). Conclusion In contrast to static PET based SUVmax, both subcutaneous and in situ inflammations and malignancies can be differentiated via dynamic FDG-PET based Ki. Moreover, Values of influx rate constant Ki from compartmental modeling can offer an assessment for inflammations at different locations of the body, which also implies further validation is necessary before the replacement of in situ inflammation with its subcutaneous counterpart in animal experiments.

Simulation study of a high‐performance brain PET system with dodecahedral geometry

Purpose In brain imaging, the spherical PET system achieves the highest sensitivity when the solid angle is concerned. However, it is not practical. In this work, we designed an alternative sphere‐like scanner, the dodecahedral scanner, which has a high sensitivity in imaging and a high feasibility to manufacture. We simulated this system and compared the performance with a few other dedicated brain PET systems. Methods Monte Carlo simulations were conducted to generate data of the dedicated brain PET system with the dodecahedral geometry (11 regular pentagon detectors). The data were then reconstructed using the in‐house developed software with the fully three‐dimensional maximum‐likelihood expectation maximization (3D‐MLEM) algorithm. Results Results show that the proposed system has a high‐sensitivity distribution for the whole field of view (FOV). With a depth‐of‐interaction (DOI) resolution around 6.67 mm, the proposed system achieves the spatial resolution of 1.98 mm. Our simulation study also shows that the proposed system improves the image contrast and reduces noise compared with a few other dedicated brain PET systems. Finally, simulations with the Hoffman phantom show the potential application of the proposed system in clinical applications. Conclusions In conclusion, the proposed dodecahedral PET system is potential for widespread applications in high‐sensitivity, high‐resolution PET imaging, to lower the injected dose.

Design of a short nonuniform acquisition protocol for quantitative analysis in dynamic cardiac SPECT imaging — a retrospective 123I‐MIBG animal study

Purpose Our previous works have found that quantitative analysis of 123I‐MIBG kinetics in the rat heart with dynamic single‐photon emission computed tomography (SPECT) offers the potential to quantify the innervation integrity at an early stage of left ventricular hypertrophy. However, conventional protocols involving a long acquisition time for dynamic imaging reduce the animal survival rate and thus make longitudinal analysis difficult. The goal of this work was to develop a procedure to reduce the total acquisition time by selecting nonuniform acquisition times for projection views while maintaining the accuracy and precision of estimated physiologic parameters. Method Taking dynamic cardiac imaging with 123I‐MIBG in rats as an example, we generated time activity curves (TACs) of regions of interest (ROIs) as ground truths based on a direct four‐dimensional reconstruction of experimental data acquired from a rotating SPECT camera, where TACs represented as the coefficients of B‐spline basis functions were used to estimate compartmental model parameters. By iteratively adjusting the knots (i.e., control points) of B‐spline basis functions, new TACs were created according to two rules: accuracy and precision. The accuracy criterion allocates the knots to achieve low relative entropy between the estimated left ventricular blood pool TAC and its ground truth so that the estimated input function approximates its real value and thus the procedure yields an accurate estimate of model parameters. The precision criterion, via the D‐optimal method, forces the estimated parameters to be as precise as possible, with minimum variances. Based on the final knots obtained, a new protocol of 30 min was built with a shorter acquisition time that maintained a 5% error in estimating rate constants of the compartment model. This was evaluated through digital simulations. Results The simulation results showed that our method was able to reduce the acquisition time from 100 to 30 min for the cardiac study of rats with 123I‐MIBG. Compared to a uniform interval dynamic SPECT protocol (1 s acquisition interval, 30 min acquisition time), the newly proposed protocol with nonuniform interval achieved comparable (Symbol and Symbol, P = 0.5745 for Symbol and P = 0.0604 for Symbol) or better (Distribution Volume, DV, P = 0.0004) performance for parameter estimates with less storage and shorter computational time. Symbol. No caption available. Symbol. No caption available. Symbol. No caption available. Symbol. No caption available. Conclusion In this study, a procedure was devised to shorten the acquisition time while maintaining the accuracy and precision of estimated physiologic parameters in dynamic SPECT imaging. The procedure was designed for 123I‐MIBG cardiac imaging in rat studies; however, it has the potential to be extended to other applications, including patient studies involving the acquisition of dynamic SPECT data.

Progresses in designing a high-sensitivity dodecahedral PET for brain imaging

Dedicated brain PET has the potential to achieve better sensitivity, spatial resolution and image quality than conventional whole body PET cameras in brain tomography. Spherical PET (S-PET), that can achieve a higher geometrical sensitivity and a lower parallax error than conventional cylindrical ring PET scanners, is a good candidate for high performance dedicated brain imaging. Our simulation studies show that the S-PET has a local geometric efficiency 2.7 times higher than the 30cm cylinder PET, and 19 times higher than the 76cm cylinder PET in the zones around the cerebral cortex. However, it is very challenging to design detectors with curved surface for S-PET. Convex polyhedron PETs, such as dodecahedral PET, is easier to build and has the potential to achieve performances equivalent to those of S-PETs. We are building a high-sensitivity dodecahedron PET for brain imaging. In this paper, we report our progresses in: (1) sensitivity simulation studies, (2) design, simulation and fabrication of a pentagon-shaped detector module, (3) design and fabrication of the mechanic gantry, and (4) image reconstruction.

A simulation study comparing different pixel sizes of CZT detectors combined with pitch-matched collimators for SPECT imaging

Pixelated direct-conversion detectors are poised to replace scintillators in single photon emission computed tomography (SPECT) cameras with pixel-matching parallel-hole collimators, to compensate for lost resolution while yielding higher sensitivity. However, smaller pixel sizes to improve resolution leads to increased inter-crystal signal sharing, resulting in a drop in sensitivity. It is, therefore, evident that the pixel pitch of the detector plays a prime role in tuning the trade-off between sensitivity and resolution. Our objective in this simulation study is to determine an acceptable combination of detector pixel and collimator hole pitch for improved performance of the SPECT gamma camera in detecting small lesions. Using the Geant4 Monte Carlo toolkit, a family of CZT detectors, with varying pixel size, mounted with corresponding parallel-hole pitch-matched collimators were simulated. For the entire study, the septal thickness of the collimators was fixed at 0.16 mm keeping in line with the Siemens LEHR collimator specifications, and the ratio of hole length to hole pitch was maintained at 16 to obtain approximately 7 mm planar collimator resolution when the source-to-collimator distance is 10 cm. From the simulations, the figures of merit such as sensitivity, resolution, signal sharing fraction (SSF) have been compared for pixel size ranging from 1 to 2 mm in steps of 0.2 mm. Recovery ratio calculated from a hot rod phantom filled with Tc-99m helped ascertain the optimum pixel design. Based on observations from the Monte Carlo simulations, we concluded that smaller pixels do not necessarily imply improved performance since they are more susceptible to lateral signal sharing. Pitch-matched collimators with larger holes exhibited higher detection efficiency but with deteriorated resolution, supporting our conclusion. The best recovery ratio was attained at 1.4 mm pixel pitch with matching collimator pitch and 19.84 mm hole length.

Dual-time point and TAC analysis of FDG-PET imaging to differentiate malignancies from inflammation in mouse models

The aim of this study is to evaluate the potential of both the Dual-time point PET imaging and the Time Activity Curve (TAC) in dynamic PET imaging in differentiating tumor and inflammation (both subcutaneous and in situ). Methods: Dynamic FDG-PET was performed on 23 female mice inoculated with tumor or inflammation subcutaneously or inside the lung. Values of Retention Index (RI) as a metric in the dual-time analysis and curvature of TAC were obtained. Results: Values of RI showed no significant difference between various experimental groups, while values of curvature showed significant difference between subcutaneous tumor and inflammation, tumors at different locations (subcutaneous and in situ), and inflammations at different locations (subcutaneous and in situ). Conclusions: Dual-time analysis requires more work in deciding the two time points. Compared to the dual-point analysis with current settings, analyzing the shape of TAC has a higher sensitivity for differentiating malignancies from inflammations.

Optimized acquisition protocols for dynamic cardiac SPECT imaging of rats with 123I-MIBG

Our previous work in dynamic cardiac SPECT imaging of rats with 123I-MIBG showed that with a slow rotation camera (dual head acquisition with 90s per rotation and a 1s acquisition interval at each angle) we could accurately obtain the time activity curves (TACs) and estimate compartmental model parameters. However, the long acquisition time (usually exceeding 60 rotations) limits the throughput and the animal survival rate. The short acquisition interval (1s) can result in the poor photon statistics which increases the variance of the TACs even though it reduce the bias of the TACs. In this study, we tried to shorten the whole acquisition time, optimize the acquisition time interval at each projection view adaptively, while maintaining the estimation accuracy of the kinetic parameters through computer simulations studies. First, the original blood pool TAC (bTAC) was obtained by averaging the bTACs of 5 WKY rats acquired previously. The tissue TAC (tTAC) was the two-tissue compartmental model output with pre-defined kinetic parameters and the original bTAC as the input. Then we cut off the first n segments of 2s, 5s, 10s, 20s and 30s in the bTAC to mimic the acquisition intervals during the acquisition. The cut-off portions were extrapolated to form new bTACs. The relative entropy of the new bTAC and the original bTAC was calculated to decide the max segment number n that could be tolerable. The same segments were also cut off in the tTAC. Finally, the resultant bTACs and tTACs were truncated with acquisition lengths of 1.5, 3, 4.5, 9, 18, 36, and 72 mins and fit to the two-tissue compartment model to estimate the kinetic parameters (K1,k2,k3,k4). The Distribution Volume (DV) was calculated from the kinetic parameters. The percentage error (PE) between the estimated parameters and pre-defined parameters were calculated. The results showed that, to match the PE with the original protocol, the kinetic parameter K1 could be estimated with an acquisition time of 90s with non-uniform acquisition protocols of 2s. The DV could be estimated with an acquisition time of 180s with non-uniform acquisition protocols of 5s.

Evaluation of sympathetic nervous system function in normal and spontaneously hypertensive rat hearts with dynamic SPECT imaging

123I-MIBG is an analogue of the norepinephrine uptake in the presynaptic portion of the sympathetic neurons that innervate the heart. The abnormal metabolism of 123I-MIBG is used to diagnose potential heart failure. This paper quantitatively compares the metabolism of 123I-MIBG in SHRs and Wistar-Kyoto normal rats. Dynamic projection data were acquired for 100 min in 1-sec time frames with an angular step of 2 degrees per frame on a dual-head GE Millennium VG Hawkeye SPECT/CT scanner equipped with custom pinhole collimators. The time activity curves of radiotracer in blood pool and myocardium were extracted from the projections through the spatiotemporal dimension reduction method developed previously for dynamic cardiac SPECT reconstruction. The kinetic parameters were estimated by fitting the estimate time activity curves to a compartmental model. The distribution volumes (DVs) of the radiotracer in the myocardium were calculated. The DV of SHRs was greater (p=2.4E-4) than that of WKY normal rats, indicating the development of cardiac hypertrophy in the SHRs.