Direct imaging of functional networks

Targeted partial reconstruction for real-time fMRI with arbitrary trajectories

PURPOSE

A partial image reconstruction formalism is introduced for the targeted extraction of real-time feedback from arbitrary trajectories when full image reconstruction in real time is computationally too demanding.

METHODS

Explicit calculation and storage of linear combinations of lines of the reconstruction matrix by an incomplete basis change in spatial coordinates lead to translation of the expensive full reconstruction from a frame-wise application to a region of interest (ROI)-wise application. This step is independent from signal data and can be executed before the experiment. Subsequently, the results of the sum over fully reconstructed voxels can be evaluated directly. Data from a high-speed fMRI acquisition was used to investigate the targeted partial reconstruction of a functional ROI atlas, incorporating an intravolume dephasing correction. The same data and ROIs were used for a comparison of the time series obtained with those obtained from already existing methods for compartment-wise reconstruction. To examine real-time feasibility, the reconstruction was implemented and tested for online reconstruction performance.

RESULTS

The reconstruction yields results that are virtually identical to the standard reconstruction (i.e., the magnitude sums over the ROIs), with negligible discrepancies even after termination of the conjugate gradient algorithm at a feasible number of iterations. Notably, more discrepancies arise with existing compartment-wise reconstructions. The online real-time implementation evaluated 1 ROI within 2.8 ms in the case of a highly parallel 3D whole brain acquisition.

CONCLUSION

The high reconstruction fidelity and speed are satisfying for the exemplary application of real-time functional feedback using a highly parallel 3D whole brain acquisition.

Recovering task fMRI signals from highly under-sampled data with low-rank and temporal subspace constraints

Recent developments in highly accelerated fMRI data acquisition have employed low-rank and/or sparsity constraints for image reconstruction, as an alternative to conventional, time-independent parallel imaging. When under-sampling factors are high or the signals of interest are low-variance, however, functional data recovery can be poor or incomplete. We introduce a method for improving reconstruction fidelity using external constraints, like an experimental design matrix, to partially orient the estimated fMRI temporal subspace. Combining these external constraints with low-rank constraints introduces a new image reconstruction model that is analogous to using a mixture of subspace-decomposition (PCA/ICA) and regression (GLM) models in fMRI analysis. We show that this approach improves fMRI reconstruction quality in simulations and experimental data, focusing on the model problem of detecting subtle 1-s latency shifts between brain regions in a block-design task-fMRI experiment. Successful latency discrimination is shown at acceleration factors up to R = 16 in a radial-Cartesian acquisition. We show that this approach works with approximate, or not perfectly informative constraints, where the derived benefit is commensurate with the information content contained in the constraints. The proposed method extends low-rank approximation methods for under-sampled fMRI data acquisition by leveraging knowledge of expected task-based variance in the data, enabling improvements in the speed and efficiency of fMRI data acquisition without the loss of subtle features.

Direct estimation of 17 O MR images (DIESIS) for quantification of oxygen metabolism in the human brain with partial volume correction

PURPOSE

To provide a data post-processing method that corrects for partial volume effects (PVE) and fast <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:mrow><mml:msubsup><mml:mi>T</mml:mi> <mml:mn>2</mml:mn> <mml:mo>*</mml:mo></mml:msubsup> </mml:mrow> </mml:math> decay in dynamic ¹⁷ O MRI for the mapping of cerebral metabolic rates of oxygen consumption (CMRO₂ ).

METHODS

CMRO₂ is altered in neurodegenerative diseases and tumors and can be measured after ¹⁷ O gas inhalation using dynamic ¹⁷ O MRI. CMRO₂ quantification is difficult because of PVE. To correct for PVE, a direct estimation of the MR images (DIESIS) method is proposed and used in 4 dynamic ¹⁷ O MRI data sets of a healthy volunteer acquired on a 3T MRI system. With DIESIS, ¹⁷ O MR signal time curves in selected regions were directly estimated based on parcellation of a coregistered ¹ H MPRAGE image.

RESULTS

Profile likelihood analysis of the DIESIS method showed identifiability of CMRO₂ . In white matter (WM), DIESES reduced CMRO₂ from 0.97 ± 0.25 µmol/g_tissue /min with Kaiser-Bessel gridding reconstruction to 0.85 ± 0.21 µmol/g_tissue /min, whereas in gray matter (GM) it increases from 1.3 ± 0.31 µmol/g_tissue /min to 1.86 ± 0.36 µmol/g_tissue /min; both values are closer to the literature values from the ¹⁵ O-PET studies.

CONCLUSION

DIESIS provided an increased separation of CMRO₂ values in GM and WM brain regions and corrected for partial volume effects in ¹⁷ O-MRI inhalation experiments. DIESIS could also be applied to more heterogeneous tissues such as glioblastomas if subregions of the tumor can be represented as additional parcels.

Time-efficient measurement of multi-phase arterial spin labeling MR signal in white matter

White matter (WM) perfusion has great potential as a physiological biomarker in many neurological diseases. Although it has been demonstrated previously that arterial spin labeling magnetic resonance imaging (ASL-MRI) enables the detection of the perfusion-weighted signal in most voxels in WM, studies of cerebral blood flow (CBF) in WM by ASL-MRI are relatively scarce because of its particular challenges, such as significantly lower perfusion and longer arterial transit times relative to gray matter (GM). Recently, ASL with a spectroscopic readout has been proposed to enhance the sensitivity for the measurement of WM perfusion. However, this approach suffers from long acquisition times, especially when acquiring multi-phase ASL datasets to improve CBF quantification. Furthermore, the potential increase in the signal-to-noise ratio (SNR) by spectroscopic readout compared with echo planar imaging (EPI) readout has not been proven experimentally. In this study, we propose the use of time-encoded pseudo-continuous ASL (te-pCASL) with single-voxel point-resolved spectroscopy (PRESS) readout to quantify WM cerebral perfusion in a more time-efficient manner. Results are compared with te-pCASL with a conventional EPI readout for both WM and GM perfusion measurements. Perfusion measurements by te-pCASL PRESS and conventional EPI showed no significant difference for quantitative WM CBF values (Student's t-test, p = 0.19) or temporal SNR (p = 0.33 and p = 0.81 for GM and WM, respectively), whereas GM CBF values (p = 0.016) were higher using PRESS than EPI readout. WM CBF values were found to be 18.2 ± 7.6 mL/100 g/min (PRESS) and 12.5 ± 5.5 mL/100 g/min (EPI), whereas GM CBF values were found to be 77.1 ± 11.2 mL/100 g/min (PRESS) and 53.6 ± 9.6 mL/100 g/min (EPI). This study demonstrates the feasibility of te-pCASL PRESS for the quantification of WM perfusion changes in a highly time-efficient manner, but it does not result in improved temporal SNR, as does traditional te-pCASL EPI, which remains the preferred option because of its flexibility in use.