Rigid motion-corrected magnetic resonance fingerprinting

PURPOSE

Develop a method for rigid body motion-corrected magnetic resonance fingerprinting (MRF).

METHODS

MRF has shown some robustness to abrupt motion toward the end of the acquisition. Here, we study the effects of different types of rigid body motion during the acquisition on MRF and propose a novel approach to correct for this motion. The proposed method (MC-MRF) follows 4 steps: (1) sliding window reconstruction is performed to produce high-quality auxiliary dynamic images; (2) rotation and translation motion is estimated from the dynamic images by image registration; (3) estimated motion is used to correct acquired k-space data with corresponding rotations and phase shifts; and (4) motion-corrected data are reconstructed with low-rank inversion. MC-MRF was validated in a standard T1 /T2 phantom and 2D in vivo brain acquisitions in 7 healthy subjects. Additionally, the effect of through-plane motion in 2D MC-MRF was investigated.

RESULTS

Simulation results show that motion in MRF can introduce artifacts in T1 and T2 maps, depending when it occurs. MC-MRF improved parametric map quality in all phantom and in vivo experiments with in-plane motion, comparable to the no-motion ground truth. Reduced parametric map quality, even after motion correction, was observed for acquisitions with through-plane motion, particularly for smaller structures in T2 maps.

CONCLUSION

Here, a novel method for motion correction in MRF (MC-MRF) is proposed, which improves parametric map quality and accuracy in comparison to no-motion correction approaches. Future work will include validation of 3D MC-MRF to enable also through-plane motion correction.

Development of a Dedicated Rebinner with Rigid Motion Correction for the mMR PET/MR Scanner, and Validation in a Large Cohort of 11C-PIB Scans

Head motion occurring during brain PET studies leads to image blurring and to bias in measured local quantities. The objective of this work was to implement a correction method for PET data acquired with the mMR synchronous PET/MR scanner. Methods: A list-mode-based motion-correction approach has been designed. The developed rebinner chronologically reads the recorded events from the Siemens list-mode file, applies the estimated geometric transformations, and frames the detected counts into sinograms. The rigid-body motion parameters were estimated from an initial dynamic reconstruction of the PET data. We then optimized the correction for ¹¹C-Pittsburgh compound B (¹¹C-PIB) scans using simulated and actual data with well-controlled motion. Results: An efficient list-mode-based motion correction approach has been implemented, fully optimized, and validated using simulated and actual PET data. The average spatial resolution loss induced by inaccuracies in motion parameter estimates and by the rebinning process was estimated to correspond to a 1-mm increase in full width at half maximum with motion parameters estimated directly from the PET data with a temporal frequency of 20 s. The results show that the rebinner can be safely applied to the ¹¹C-PIB scans, allowing almost complete removal of motion-induced artifacts. The application of the correction method to a large cohort of ¹¹C-PIB scans led to the following observations: first, that more than 21% of the scans were affected by motion greater than 10 mm (39% for subjects with Mini-Mental State Examination scores below 20), and second, that the correction led to quantitative changes in Alzheimer-specific cortical regions of up to 30%. Conclusion: The rebinner allows accurate motion correction at a cost of minimal resolution reduction. Application of the correction to a large cohort of ¹¹C-PIB scans confirmed the necessity of systematically correcting for motion to obtain quantitative results.

Optimizing the frame duration for data-driven rigid motion estimation in brain PET imaging

PURPOSE

Data-driven rigid motion estimation for PET brain imaging is usually performed using data frames sampled at low temporal resolution to reduce the overall computation time and to provide adequate signal-to-noise ratio in the frames. In recent work it has been demonstrated that list-mode reconstructions of ultrashort frames are sufficient for motion estimation and can be performed very quickly. In this work we take the approach of using image-based registration of reconstructions of very short frames for data-driven motion estimation, and optimize a number of reconstruction and registration parameters (frame duration, MLEM iterations, image pixel size, post-smoothing filter, reference image creation, and registration metric) to ensure accurate registrations while maximizing temporal resolution and minimizing total computation time.

METHODS

Data from 18 F-fluorodeoxyglucose (FDG) and 18 F-florbetaben (FBB) tracer studies with varying count rates are analyzed, for PET/MR and PET/CT scanners. For framed reconstructions using various parameter combinations interframe motion is simulated and image-based registrations are performed to estimate that motion.

RESULTS

For FDG and FBB tracers using 4 × 105 true and scattered coincidence events per frame ensures that 95% of the registrations will be accurate to within 1 mm of the ground truth. This corresponds to a frame duration of 0.5-1 sec for typical clinical PET activity levels. Using four MLEM iterations with no subsets, a transaxial pixel size of 4 mm, a post-smoothing filter with 4-6 mm full width at half maximum, and averaging two or more frames to create the reference image provides an optimal set of parameters to produce accurate registrations while keeping the reconstruction and processing time low.

CONCLUSIONS

It is shown that very short frames (≤1 sec) can be used to provide accurate and quick data-driven rigid motion estimates for use in an event-by-event motion corrected reconstruction.

Improved motion correction in PROPELLER by using grouped blades as reference

PURPOSE

To develop a robust reference generation method for improving PROPELLER (Periodically Rotated Overlapping ParallEL Lines with Enhanced Reconstruction) reconstruction.

MATERIALS AND METHODS

A new reference generation method, grouped-blade reference (GBR), is proposed for calculating rotation angle and translation shift in PROPELLER. Instead of using a single-blade reference (SBR) or combined-blade reference (CBR), our method classifies blades by their relative correlations and groups similar blades together as the reference to prevent inconsistent data from interfering the correction process. Numerical simulations and in vivo experiments were used to evaluate the performance of GBR for PROPELLER, which was further compared with SBR and CBR in terms of error level and computation cost.

RESULTS

Both simulation and in vivo experiments demonstrate that GBR-based PROPELLER provides better correction for random motion or bipolar motion comparing with SBR or CBR. It not only produces images with lower error level but also needs less iteration steps to converge.

CONCLUSION

A grouped-blade for reference selection was investigated for PROPELLER MRI. It helps to improve the accuracy and robustness of motion correction for various motion patterns.

Rigid motion tracking using moments of inertia in TOF-PET brain studies

A data-driven method is proposed for rigid motion estimation directly from time-of-flight (TOF)-positron emission tomography (PET) emission data. Rigid motion parameters (translations and rotations) are estimated from the first and second moments of the emission data masked in a spherical volume. The accuracy of the method is analyzed on 3D analytical simulations of the PET-SORTEO brain phantom, and subsequently tested on¹⁸F-FDG as well as¹¹C-PIB brain datasets acquired on a TOF-PET/CT scanner. The estimated inertia-based motion is later compared to rigid motion parameters obtained by directly registering the short frame backprojections. We find that the method provides sub mm/degree accuracies for the estimated rigid motion parameters for counts corresponding to typical 0.5 s, 1 s, and 2 s¹⁸F-FDG brain scans, with the current TOF resolutions clinically available. The method provides robust motion estimation for different types of patient motion, most notably for a continuous patient motion case where conventional frame-based approaches which rely on little to no intra-frame motion of short time intervals could fail. The method relies on the detection of stable eigenvectors for accurate motion estimation, and a monitoring of this condition can reveal time-frames where the motion estimation is less accurate, such as in dynamic PET studies.