Synopsis

Most diffusion MRI is today performed with the so-called pulsed gradient spin echo (PGSE) method, which encodes for diffusion using two gradient pulses. This method is sensitive to cellularity of tumours, orientation of white matter tracts, and microstructure features such as axon density and cell sizes. However, the PGSE method is fundamentally limited in several ways. This talk will pinpoint these limitations and show how novel gradient waveforms can overcome them.

Background

Diffusion MRI use endogenous molecules to non-invasively probe the tissue microstructure [Nilsson et al, 2013a]. The majority of all present dMRI studies are performed with the so-called pulsed gradient spin echo (PGSE) method, which encodes for diffusion using two gradient pulses. This method, which was invented more than half a century ago [Stejskal and Tanner, 1965], yields diffusion-weighted images sensitive to for example cellularity of tumours [Chen et al, 2013] and the orientation of white matter tracts [Basser et al, 1994]. By posing and solving an inverse problem, microstructural tissue properties such as axon density and cell sizes can also be estimated from the image data. However, the PGSE method is fundamentally limited in several ways. This talk will pinpoint these limitations and show how novel gradient waveforms can overcome them. Examples of limitations are provided below:

• Pore mapping with q-space MRI. The autocorrelation function of the pore volume can be computed by an inverse Fourier transform of the signal. Unfortunately, the autocorrelation operation intrinsic to the method smears out fine details of the pore function. This limitation can be overcome by acquiring data with long-short gradients [Kuder and Laun, 2015].

• Mapping microscopic anisotropy. With PGSE, effects of microscopic anisotropy are invariably entangled with orientation dispersion [Mitra et al, 1995]. Methods such as NODDI has been devised to disentangle these properties [Zhang et al, 2012], although it has to rely on strong model assumptions. To separate these effects in a data-driven way, gradient waveforms that permit tensor-valued diffusion encoding can be used [Lampinen et al, 2017].

• Measuring the perfusion fraction. The IVIM method takes advantage of the high pseudo-diffusion of capillary blood to estimate the perfusion fraction [Le Bihan et al, 1986]. However, the estimated perfusion fraction is highly sensitive to noise due to the ill-posed nature of the inverse problem in IVIM. By acquiring data with variable level of flow compensation, more accurate estimates of the perfusion fraction can be obtained [Wetscherek et al 2015, Ahlgren et al, 2016].

• Estimating axon diameters. Accurate estimation of axon diameters require data to be collected with the maximal possible b-value, however, in the presence of orientation dispersion this leads to poor SNR. Oscillating gradients offers an alternative that permits the same sensitivity to the diameter but at lower b-values, thus retaining SNR [Drobnjak et al, 2015; Nilsson et al, 2017].

• Detecting water exchange. Effects of water exchange acts on the signal in opposite ways to effects of restricted diffusion with PGSE [Nilsson et al, 2013a]. Thus, these effects are difficult to disentangle which leads to noisy parameter estimates. Methods based on double-diffusion encoding, such as filter exchange imaging, can overcome this problem by constructing a protocol with specific sensitivity to exchange [Lasic et al 2011, Nilsson et al 2013b].

Acknowledgements

References

Ahlgren, André, et al. "Quantification of microcirculatory parameters by joint analysis of flow-compensated and non-flow-compensated intravoxel incoherent motion (IVIM) data." NMR in Biomedicine 29.5 (2016): 640-649.

Basser, Peter J., James Mattiello, and Denis LeBihan. "MR diffusion tensor spectroscopy and imaging." Biophysical journal 66.1 (1994): 259-267.

Chen, Lihua, et al. "The correlation between apparent diffusion coefficient and tumor cellularity in patients: a meta-analysis." PloS one 8.11 (2013): e79008.

Drobnjak, Ivana, et al. "PGSE, OGSE, and sensitivity to axon diameter in diffusion MRI: Insight from a simulation study." Magnetic resonance in medicine (2015).

Kuder, Tristan Anselm, and Frederik Bernd Laun. "Effects of pore-size and shape distributions on diffusion pore imaging by nuclear magnetic resonance." Physical Review E 92.2 (2015): 022706.

Lampinen, Björn, et al. "Neurite density imaging versus imaging of microscopic anisotropy in diffusion MRI: A model comparison using spherical tensor encoding." NeuroImage 147 (2017): 517-531.

Lasic, Samo, et al. "Apparent exchange rate mapping with diffusion MRI." Magnetic resonance in medicine 66.2 (2011): 356-365.

Le Bihan, Denis, et al. "Separation of diffusion and perfusion in intravoxel incoherent motion MR imaging." Radiology 168.2 (1988): 497-505.

Mitra, Partha P. "Multiple wave-vector extensions of the NMR pulsed-field-gradient spin-echo diffusion measurement." Physical Review B 51.21 (1995): 15074.

Nilsson, Markus, et al. "The role of tissue microstructure and water exchange in biophysical modelling of diffusion in white matter." Magnetic Resonance Materials in Physics, Biology and Medicine 26.4 (2013a): 345-370.

Nilsson, Markus, et al. "Noninvasive mapping of water diffusional exchange in the human brain using filter-exchange imaging." Magnetic resonance in medicine 69.6 (2013b) 1572-1580.

Nilsson, Markus, et al. “Resolution limit of cylinder diameter estimation by diffusion MRI: The impact of gradient waveform and orientation dispersion”. NMR Biomed (2017).

Stejskal, Edward O., and John E. Tanner. "Spin diffusion measurements: spin echoes in the presence of a time-dependent field gradient." The journal of chemical physics 42.1 (1965): 288-292.

Wetscherek, Andreas, Bram Stieltjes, and Frederik Bernd Laun. "Flow-compensated intravoxel incoherent motion diffusion imaging." Magnetic resonance in medicine 74.2 (2015): 410-419. Zhang, Hui, et al. "NODDI: practical in vivo neurite orientation dispersion and density imaging of the human brain." Neuroimage 61.4 (2012): 1000-1016.