Correction of hysteretic respiratory motion in SPECT myocardial perfusion imaging: Simulation and patient studies

Abstract

Purpose: Amplitude-based respiratory gating is known to capture the extent of respiratory motion (RM) accurately but results in residual motion in the presence of respiratory hysteresis. In our previous study, we proposed and developed a novel approach to account for respiratory hysteresis by applying the Bouc-Wen (BW) model of hysteresis to external surrogate signals of anterior/posterior motion of the abdomen and chest with respiration. In this work, using simulated and clinical SPECT myocardial perfusion imaging (MPI) studies, we investigate the effects of respiratory hysteresis and evaluate the benefit of correcting it using the proposed BW model in comparison with the abdomen signal typically employed clinically.

Methods: The MRI navigator data acquired in free-breathing human volunteers were used in the specially modified 4D NCAT phantoms to allow simulating three types of respiratory patterns: monotonic, mild hysteresis, and strong hysteresis with normal myocardial uptake, and perfusion defects in the anterior, lateral, inferior, and septal locations of the mid-ventricular wall. Clinical scans were performed using a Tc-99m sestamibi MPI protocol while recording respiratory signals from thoracic and abdomen regions using a visual tracking system (VTS). The performance of the correction using the respiratory signals was assessed through polar map analysis in phantom and 10 clinical studies selected on the basis of having substantial RM.

Results: In phantom studies, simulations illustrating normal myocardial uptake showed significant differences (P < 0.001) in the uniformity of the polar maps between the RM uncorrected and corrected. No significant differences were seen in the polar map uniformity across the RM corrections. Studies simulating perfusion defects showed significantly decreased errors (P < 0.001) in defect severity and extent for the RM corrected compared to the uncorrected. Only for the strong hysteretic pattern, there was a significant difference (P < 0.001) among the RM corrections. The errors in defect severity and extent for the RM correction using abdomen signal were significantly higher compared to that of the BW (severity = -4.0%, P < 0.001; extent = -65.4%, P < 0.01) and chest (severity = -4.1%, P < 0.001; extent = -52.5%, P < 0.01) signals. In clinical studies, the quantitative analysis of the polar maps demonstrated qualitative and quantitative but not statistically significant differences (P = 0.73) between the correction methods that used the BW signal and the abdominal signal.

Conclusions: This study shows that hysteresis in respiration affects the extent of residual motion left in the RM-binned data, which can impact wall uniformity and the visualization of defects. Thus, there appears to be the potential for improved accuracy in reconstruction in the presence of hysteretic RM with the BW model method providing a possible step in the direction of improvement.

Keywords: SPECT; cardiac imaging; hysteresis; respiratory motion.

Figure 1

Schematic diagram illustrates the signal acquisitions and their usage in the simulation process. Simultaneous acquisition of internal (MRI navigator) and external (VTS) motion data from free‐breathing volunteers were employed to generate realistic organ RM in NCAT phantoms simulating SPECT MPI, and correct for this motion by employing the external respiratory signals in amplitude‐based motion correction strategies. [Color figure can be viewed at wileyonlinelibrary.com]

Figure 2

Plots for the MRI navigator measured internal motion on the heart and diaphragm for the monotonic, mild, and strong hysteresis respiratory patterns simulated in the phantom studies. Plotted data portray individual navigator measurements of the heart and diaphragm position for each of the 300 measured time points. Notice that as the extent of hysteresis increased the heart takes on increasingly different positions during inspiration (red points online) and expiration (blue points online) for the same diaphragm location. [Color figure can be viewed at wileyonlinelibrary.com]

Figure 3

Schematic illustrations show the short‐axis (left) and long‐axis (right) views of the left ventricle wall. The shaded areas represent the locations of the perfusion defects simulated on the anterior (${\mathit{\theta }}_{center}=0^{\circ }$), lateral (${\mathit{\theta }}_{center}={90}^{\circ }$), inferior (${\mathit{\theta }}_{center}={180}^{\circ }$), and septal (${\mathit{\theta }}_{center}={270}^{\circ }$) walls with defect span (Δ _θ )$={60}^{\circ }$, and defect size $({\mathrm{\Delta }}_z)=3cm$.

Figure 4

Block diagram illustrates the experimental design of the simulation study. The simulations were performed for the three different respiratory patterns. Each respiratory pattern was simulated for a healthy heart (Normal) and a heart with perfusion abnormality (Perfusion Defect) separately in the mid‐anterior, lateral, inferior, and septal regions of the left ventricle.

Figure 5

Plots of respiratory motion as estimated by registration to the reference bin (bin #5) for (a) monotonic, (b) mild hysteretic and (c) strong hysteretic respiratory patterns obtained from the reconstructed NCAT slices of amplitude‐binned projections using the surrogate respiratory signals from the abdomen, chest, and the BW model versus the heart navigator (true). Shown are the mean and standard deviation for the 10 noise realizations. Also, shown are the correlation coefficients (r) of the estimates of the respiratory surrogate signals with the true. [Color figure can be viewed at wileyonlinelibrary.com]

Figure 6

Three rows of box plots for estimated respiratory motion versus respiratory bin number for the monotonic (a), mild, (b) and strong, (c) hysteretic respiratory patterns. From left to right, the motion was estimated using the MRI‐navigator‐derived true respiratory signal for heart motion with respiration, the VTS‐derived abdomen signal, the VTS‐derived chest signal, and the Bouc–Wen‐derived signal. On each box, the central mark is the median, the edges of the box are the 25th and 75th percentiles, the whiskers are the most extreme data points, and the ‘+’ symbol are the outliers. [Color figure can be viewed at wileyonlinelibrary.com]

Figure 7

Rows (a), (b), and (c) consist of polar maps for monotonic, mild, and strong hysteretic respiratory patterns from the normal heart phantom reconstructions. Polar maps derived from the uncorrected and respiratory motion corrected reconstructions using the abdomen (Abd), chest, and the Bouc–Wen model (BW), and the heart navigator (True) signals are shown. Uniformity values for the corresponding polar maps are given as mean ± standard deviation. Note that the apical cooling is expected in the NCAT phantom due to thinning of the wall in this location. Thus, a perfectly uniform polar map is not to be expected and the results for usage of the true signal should be taken as the truth to compare against. [Color figure can be viewed at wileyonlinelibrary.com]

Figure 8

Examples of one noise realization each for polar maps and blackout polar maps of the perfusion defect NCAT phantoms for each of the three respiratory patterns (monotonic, mild hysteresis, and strong hysteresis) and the four defect locations (a), (b), (c), and (d). Polar maps of uncorrected and corrected reconstructions for the abdomen (Abd), chest, Bouc–Wen model (BW), and heart navigator (True) used in motion correction algorithm are shown. [Color figure can be viewed at wileyonlinelibrary.com]

Figure 9

Histograms showing the average and standard deviation (error bar) of 10 noise realizations each for defect intensity and defect area (expressed as percent errors) in the uncorrected and corrected reconstructions using the respiratory signals from the abdomen, chest, and the Bouc–Wen model for the anterior (a), lateral (b), inferior (c), and septal (d) defect locations. Results are shown for the monotonic, mild, and strong hysteretic respiratory patterns. The calculated errors which are significantly different with *P < 0.05, **P < 0.01, or ***P < 0.001 by Paired Student's t‐test with Bonferroni correction are indicated. [Color figure can be viewed at wileyonlinelibrary.com]

Figure 10

(a) The magnitudes of the extent of heart motion estimated in the SI direction using the abdomen (Abd) and Bouc–Wen (BW) signals. In (b) are shown the ratios of the counts in the anterior and inferior to lateral and septal walls of the polar maps for the uncorrected, corrected using the abdomen surrogate signal, and corrected using the BW surrogate signal. Since the anterior and inferior walls are blurred more by respiratory motion than the lateral and septal walls, one would expect in the absence of true perfusion defects, this ratio of wall counts is to be restored closer to 1.0 with respiratory motion correction. Note: The asterisk (*) indicates the studies with hysteretic patterns.

Figure 11

Three short‐axis slices (top three rows) and polar maps (bottom row) of Tc‐99m sestamibi stress scans for two patients without and with RM correction using the abdomen and the Bouc–Wen signals. Patients 8 and 9 exhibited monotonic and mild hysteretic patterns, respectively. [Color figure can be viewed at wileyonlinelibrary.com]