LKB - Polarised Helium, Quantum Fluids and Solids



Methodological developments for MR imaging at low magnetic field

MRI with hyperpolarised gases – In-vitro and ex-vivo validation for lung MRI

Overview of team's activitiesAperçu des activités de l'équipe


Nuclear Magnetic Resonance, NMR, and NMR-based imaging, MRI, usually rely on detection of nuclear magnetic moments from a dense sample (e.g., of hydrogen nuclei in water or tissues) that are passively polarised by an intense magnetic field (1.5 T or higher)>. One may also use a gas (with a much lower number density) to image the airspaces within the lungs, provided that it is optically hyperpolarised prior to inhalation. In this case, a high field is not necessary any more.
Actually, operation at much lower field strength (in the mT range) may be advantageous. For instance, the local field inhomogeneities resulting from the lung micro-structures and tissue magnetic susceptibility are strongly reduced, which yields much longer lifetimes for transverse magnetisation (i.e., for the NMR signals) and, hence, can allows detection with higher signal-to-noise ratios.

Our group performs methodological developments and in vitro validation of MRI at low field (B= 1 – 6 mT) at LKB, with on-site production of hyperpolarised gas, and resorts to collaborations for high-field and/or in vivo tests.
We have designed and built a first compact MRI system (29 cm in diameter, 32 cm in length) for experiments up to 5 mT and volumes samples up to (10 cm)³.
This prototype device corresponds to a 1:4 scale model of a full-body 3D low-field scanner. It has been used for initial demonstrations of MRI of polarised Helium-3 gas in cells and ex-vivo lungs, as well as for systematic NMR studies of diffusion for a confined gas (restricted diffusion).
Beyond their specific interest, these pioneering studies were relevant for applications in pre-clinical MRI research with hyperpolarised gases – a research field which did involve, and still does, several groups worldwide.

P.J. Nacher, M. Pelissier, G. Tastevin, “Design and test of a magnet and gradient system for hyperpolarised gas lung MRI at ultra-low field”, Proc. Intl. Soc. Mag. Reson. Med. (2007) 15, 3290.

The experimental prototype has recently been rebuilt so as to obtain a more uniform magnetic field and longer signal decay times (up to several seconds). The new MRI system has been fully characterised by magnetic and NMR measurements. It is now used to develop and test innovative methods in low-field NMR and MRI, both on polarised gas and on water samples.

Recent work deals with:
• tests of gradient-free imaging (an emerging technique in which the spatial phase encoding of the NMR signal is achieved by the RF excitation field, rather than ced by static field gradients)
• investigation of response to very short RF pulses (NMR pulse duration is reduced to a few time periods, so that the RF field amplitude must be proportionally increased to achieve the targeted flip angle).
The upgraded MRI system has also been used, for instance, for the investigation of spontaneous maser onsets in hyperpolarised ³He gas at low or moderate pressure (i.e., in a system with negligible dipolar couplings) – a series of measurements complementary to those performed in hyperpolarised liquid ³He–⁴He mixtures (L3- and M2-level internships, 2013-2014).

V.V. Kuzmin, Dan Moinard, P.-J. Nacher, T.M. Salikhov, G. Tastevin, “Spontaneous NMR precession in hyperpolarised 3He”, PiNG14 – Sept. 2014


Internship projects

The gradient-free imaging scheme TRASE (TRansmit Array Spatial Encoding, TRASE), has originally been proposed in 2010 by a group in Winnipeg and demonstrated at moderate field strength (0.2 T), so far. At LKB, we have implemented it at low field for the first time and we have demonstrated successful operation both with thermally polarised water and with hyperpolarised ³He gas .

C.P. Bidinosti, G. Tastevin, P.-J. Nacher, “Tests of RF phase encoding in low-field MRI”, EUROHyperPOL meeting – June 2014.
C.P. Bidinosti, P.-J. Nacher, G. Tastevin, “Exploration of TRASE MRI at low field with hyperpolarised 3He”, PiNG14 – Sept. 2014.

A few unexpected artefacts appear on the first low-field TRASE images. We aim at finding the origin of these artefacts and wish to perform independent studies of two physical processes that may explain the observed imperfections. The search for qualitative and quantitative understanting will involve computer lattice simulations as well as new experimental tests with the MRI system.

P.-J. Nacher, G. Tastevin,C.P. Bidinosti, “Exploration of gradient-free (TRASE) MRI at low field with hyperpolarized 3He”, ESMRMB 2015

Internship project 1:
Impact of concomitant field gradients on image quality

Maxwell’s equations impose stringent constraints to non-uniform magnetic field maps.
In conventional MRI, thanks to the high field strength, one can legitimately neglect the contribution the additional static field components – the so-called {concomitant gradients} – introduced by the applied encoding gradients (which generate 1D linear variations of the Larmor frequency and allow simple image reconstruction, by inverse Fourier transform).
In contrast, at low field strength, the perturbations introduced by these (static) concomitant gradients are significant and lead to well-known image artefacts.

Internship work The student will study the impact of the concomitant field gradients on image quality in our experimental conditions. He/She will particularly focus on the additional field components associated to the non-uniform RF excitation patterns used for TRASE imaging, whose contribution remains to be thoroughly analysed.


Internship project 2:
The short-pulse limit: revisiting the Bloch-Siegert effect in NMR

Operation at low field (hence, at low NMR frequency) is prone to perturbative effects due to the counter-rotating circular component of the RF field for intense excitation pulses with a single coil.
The breakdown of the rotating wave approximation, which traditionally results in the well-known Bloch-Siegert shift, has been observed to lead in our experiments to very unusual behaviour and to specific experimental problems.

C.P. Bidinosti, G. Tastevin, P.-J. Nacher, “Achieving accurate tip angles in NMR at low magnetic field”, PiNG14 – Sept 2014.

Internship work The student will contribute, by analytical and numerical studies, to the ongoing effort to quantitatively account for the observations: large frequency shifts, skewed trajectories on the Bloch sphere, etc. He/She may perform further experimental tests and, hopefully, participate in the improvement of TRASE pulse sequences.
Both projects will provide opportunity for hands-on training in low-field NMR or MRI, through experiments on polarised gas or water samples and careful data analysis based on new numerical lattice simulations. Comparison between expectations and findings will be systematic.



PhD work

An important objective of our group is the development of new methods with improved performance for low field MRI and a precise evaluation of their pros and cons.

Application to lung imaging currently provides the driving momentum to this work. The evaluation of the potential benefits (improved time or spatial resolution, contrast, sensitivity, specificity, etc) and of the practical limitations of low-field MRI (with or without hyperpolarisation) will be based on quantitative measurements and comparison with more standard high-field results.

One challenging issue in lung MRI with gas probes lies in the impact of restricted diffusion, inside the complex multi-scale airways structure, on image quality and contrast. This impact may be quite significant, since diffusion is much faster in gas samples than in liquids or tissues.

  • On the one hand, for fast diffusion, damping of echo signals is significantly increased (especially for high k-vector spatial encoding) and spatial resolution is limited (the imprinted phase pattern gets quickly blurred).
  • On the other hand, restricted diffusion may occur (the length scale associated with fast free diffusion becoming comparable or larger than the relevant dimension of the open compartement – alveolus, alveolar bag, acinus, or bronchus, in the case of the lung) and induce significant local changes in the cumulated phase difference (during the encoding, free evolution, or reading steps, of image acquisition).

In short, fast free diffusion and/or restricted diffusion potentially affect the image quality but provide, also, a contrast contrast mechanism which may be used to obtain valuable formation on the complex topology and the large size dispersion of the lung airspaces. A quantitative assessment is needed, based on experiments with anatomical samples or with realistic fantoms, as well as on simulations with suitable 3D models.

Beyond useful niche applications for (pre)clinical research, the methodological developments and the MRI tests performed in vitro, ex vivo, and in vivo (on small animals, to start with) in our compact prototype system are expected to lead to in-depth understanding of new physical effects and to significant technical breakthroughs.

The student’s work will benefit from established collaborations with academic research teams in France, Europe, and Canada.