Advanced diffusion MRI for microstructure imaging
-
1
University College London, Centre for Medical Image Computing, United Kingdom
-
2
Champalimaud Centre for the Unknown, Champalimaud Neuroscience Programme, Portugal
Non-invasive estimation of pore size and shape is a key challenge in diffusion MRI, with potential applications to porous media [1,2] and biomedical imaging [3,4,5].
The first part of this work shows that such microstructural parameters can be estimated from diffusion MRI using a model-based approach to analyse the data [6,7]. The technique uses a geometric model of finite cylinders with gamma distributed radii to represent pores of various sizes and eccentricities. We consider both macroscopically isotropic and anisotropic substrates and we use Monte Carlo simulations to generate synthetic data. We compare the sensitivity of single and double diffusion encoding (SDE and DDE) sequences to the size distribution and eccentricity and further analyse different protocols of DDE sequences with parallel and/or perpendicular pairs of gradients. We show that explicitly accounting for size distribution is necessary for accurate microstructural parameter estimates, and a model that assumes a single size yields biased eccentricity values. We also find that SDE sequences support estimates, although DDE sequences with mixed parallel and perpendicular gradients enhance accuracy.
Motivated by a recent work from Drobnjak et al. [8], which shows that oscillating gradients improve sensitivity to axon diameter, we replace the pulsed gradients in the standard DDE sequence [9] with oscillating waveforms [10]. In simulation, we use a model of randomly oriented finite cylinders to compare the sensitivity and specificity of DDE and double-oscillating-diffusion-encoding (DODE) sequences to pore size and length for wide range of practical sequence parameters. The results show that DODE sequences with low frequency improve the sensitivity to pore diameter, while DDE sequences have higher sensitivity to pore length. As DDE and DODE sequences provide slightly different contrasts, a combination of these sequence enhances the sensitivity to microstructural features. Moreover, DODE and DDE with finite gradient pulses exhibit higher sensitivity compared to the ideal sequences with short pulses which are generally desired in DDE experiments, stressing the importance of optimising the acquisition protocol for a given application.
[1] Shemesh et al., Journal of the American Chemical Society 23 (2011), 757-80
[2] Topgaard and Söderman, Magnetic Resonance Imaging 21 (2003), 69-76
[3] Panagiotaki el al., Cancer Research , 74 (2014), 1902–1912
[4] Szczepankiewicz et al., NeuroImage 104 (2015), 241–52.
[5] Kleinnijenhuis et al., Cortex 25 (2013) 2569–82
[6] Ianus et al., Information Processing in Medical Imaging, 2015
[7] Ianus et al., NMR in Biomedicine, accepted
[8] Drobnjak et al., Magnetic Resonance in Medicine, 75 (2016), 688-700
[9] Mitra, Physical Review B, 51 (1995) 15074–15078.
[10] Shemesh et al, Proceedings of the 23rd meeting of the International Society for Magnetic Resonance in Medicine, 2015
Keywords:
Diffusion,
MRI,
dMRI,
multidimensional,
diffusion encoding
Conference:
New dimensions in diffusion encoding, Fjälkinge, Sweden, 11 Jan - 14 Jan, 2016.
Presentation Type:
Oral presentation
Topic:
New Dimensions in Diffusion Encoding
Citation:
Ianuş
A,
Drobnjak
I,
Semesh
N and
Alexander
DC
(2016). Advanced diffusion MRI for microstructure imaging.
Front. Phys.
Conference Abstract:
New dimensions in diffusion encoding.
doi: 10.3389/conf.FPHY.2016.01.00001
Copyright:
The abstracts in this collection have not been subject to any Frontiers peer review or checks, and are not endorsed by Frontiers.
They are made available through the Frontiers publishing platform as a service to conference organizers and presenters.
The copyright in the individual abstracts is owned by the author of each abstract or his/her employer unless otherwise stated.
Each abstract, as well as the collection of abstracts, are published under a Creative Commons CC-BY 4.0 (attribution) licence (https://creativecommons.org/licenses/by/4.0/) and may thus be reproduced, translated, adapted and be the subject of derivative works provided the authors and Frontiers are attributed.
For Frontiers’ terms and conditions please see https://www.frontiersin.org/legal/terms-and-conditions.
Received:
07 Jul 2016;
Published Online:
07 Jul 2016.
*
Correspondence:
Prof. Daniel C Alexander, University College London, Centre for Medical Image Computing, London, WC1E 6BT, United Kingdom, d.alexander@cs.ucl.ac.uk