Evaluation of frame-based and event-by-event motion-correction methods for awake monkey brain PET imaging

Abstract

PET imaging of nonhuman primates (NHPs) requires correction of head motion if the subjects are scanned awake and their heads are unrestrained, because the NHPs move their heads faster and more frequently than human subjects. This work focuses on designing and validating 2 motion-correction algorithms for awake NHP brain PET imaging.

Methods: Two motion-correction methods were implemented for awake NHP brain PET imaging: multiacquisition frame (MAF) and event-by-event (EBE). Motion data were acquired from an external motion-tracking device. The MAF method divides scan data into short subframes, reconstructs each subframe individually, and registers them to a reference orientation. This method suffers from residual intraframe motion and data loss when motion is large because a minimum frame duration is often required. The EBE method, previously implemented for a human brain scanner and adapted for a small-animal PET scanner in this work, eliminates intraframe motion and should have a best accuracy. We first evaluated the accuracy of both motion-correction methods with moving phantom scans. Both motion-correction methods were then applied to awake NHP brain PET studies with a gamma-aminobutyric acid A-benzodiazepine receptor ligand, (11)C-flumazenil, and the reconstructed images were compared with those from a motion-free anesthetized study.

Results: The phantom studies showed that EBE motion correction recovers the contrast (within 3%) similarly to the static study, whereas MAF motion correction using the standard algorithm setting showed a 25% reduction in contrast from the static case. In awake NHP brain PET imaging, EBE motion correction better recovers the fine structures than the MAF method, as compared with anesthetized studies.

Conclusion: The large magnitude and frequency of NHP head motion suggests that EBE motion correction with accurate externally measured motion data can noticeably alleviate image blurring due to the intraframe motion in the MAF motion-correction method.

Keywords: PET; awake; monkey; motion correction.

Figures

Fig. 1

Steps in the multiple-acquisition frame based motion correction method. The raw list-mode data are divided into subframes (step 1). The durations of the subframes are 8.6 ± 8.7 sec for the five awake NHP studies. Each subframe is reconstructed with FBP with a Ramp filter at Nyquist frequency, without attenuation or scatter corrections (step 2). A transmission image of the same subject from an anesthetized study is resliced to the awake reference orientation (step 3). The transmission image is resliced to the orientation of each subframe (step 4). Each subframe is re-reconstructed with attenuation and scatter corrections (step 5). Finally, the subframe images are transformed to the reference orientation and grouped to 5-min frames.

Fig. 2

Reconstructed images of a moving mini-Derenzo phantom study (left and center columns) and a static mini-Derenzo phantom study (right column). The top row shows the reconstructed images using FBP for (A) the moving phantom without motion correction, (B) with MAF motion correction (IMFT = 1mm, MDFT = 3s), and (C) the static phantom with the same total counts. The bottom row shows the reconstructed images using MOLAR for (D) the moving phantom without motion correction, (E) with EBE motion correction, and (F) the static phantom with the same total counts. All images are displayed in the transaxial view, with the image intensities displayed to a global scale. As denoted by the arrows, the rod structures are blurred in B compared with C, due to the residual intra-frame motion of the MAF method. For EBE, the rod structures in E give comparable intensity levels to F.

Fig. 3

Contrast recovery coefficient (CRC) calculation on the mini-Derenzo phantom. Each hot ROI contained 5 pixels of 0.95mm × 0.95mm for the 2.4 mm rods and 9 pixels for the 3.2 mm rods, and each cold ROI contained 4 pixels for the 2.4 mm rods and 5 pixels for the 3.2 mm rods. The CRC was calculated on Figs. 2B, C, E and F.

Fig. 4

Nonhuman primate (NHP) images for an awake study (A, B, D and E) ([11C]flumazenil, with data taken from 30 to 60 min postinjection) and an anesthetized study (C and F) using the same tracer with comparable counts in the same animal. The top row shows the images reconstructed with FBP of (A) an awake study without motion correction, (B) with MAF motion correction, and (C) an anesthetized study. The bottom row shows the images reconstructed with MOLAR of (D) the awake study without motion correction, (E) with EBE motion correction and (F) the anesthetized study. All images were registered to an NHP template, and are displayed on a common scale. As indicated by the arrows, the contrasts of the cortical structures are comparable in E and F, whereas some blurring is observed in B compared with C, due to the residual intra-frame motion of the MAF method.