VIEWS UNA MIRADA NUEVA CON RESONANCIA MAGNTICA DESARROLLADA

VIEWS: UNA MIRADA NUEVA CON RESONANCIA MAGNÉTICA (DESARROLLADA EN EL I 3 M) Joseba Alonso

MRILab Unique scanners and techniques pushing the limits of Magnetic Resonance Imaging https: //i 3 m-detectors. com/mrilab Dr. J. Alonso Group leader M. Murbach Senior scientist Dr. A. Ríos CEO Prof. Dr. J. M. Benlloch Institute director Dr. F. Galve Senior scientist E. Pallás Mechanical engineer P. Borreguero Physicist Dr. J. P. Rigla Researcher Dr. E. Díaz Researcher Dr. J. M. Algarín Postdoc C. Gramage Electronics engineer D. Grau Ph. D student R. Bosch Electronics engineer J. M. González Ph. D Student

Current projects HISTO-MRI Hard-tissue MRI New sequences High resolution MRI and PNS studies Dental applications and others Volumetric and hard tissue imaging + low field systems, control system development, nano-particles… 3

MRI is great… 4

MRI is great… limitations? • Image quality • • spatial resolution SNR contrast Sensitivity to different tissues • Scan time • Cost • Patient comfort What can we do to push the limits? What are the cornerstones of MRI? 5

k-space Fourier transform 6

VIEWS: Volumetric Image Encoding Without k-Space 7

VIEWS goals • Single shot, volumetric MRI (no phase-encoded directions) • High SNR • Suppressed acoustics • Arbitrary time evolution • Arbitrary spatial inhomogeneities • Engineered spatial resolution • Encoding Matrix based reconstruction (no Fourier) • No hardware changes required 8

Patent 9

Brief reminder: what is k-space? Spatial-frequency space Real space Fourier transform 10

The Fourier Transform in MRI Induced signal Fourier Transform 11

Fourier Transforms can be suboptimal 12

Not a huge surprise What are people really using? 13

VIEWS: Volumetric Image Encoding Without k-Space Reconstruction Encoding 14

Frequency modulation in 1 D Information concentrated in frequency modulation sidebands Every voxel contributes to all sidebands Bessel functions are better suited for this scenario than plane waves 15

New idea? Invert Encoding Matrix or Integrate numerous complex exponentials (not with an i 386…) 16

Non-linear-gradient people know how to reconstruct… Algebraic Reconstruction Techniques on GPUs 17

Frequency modulation in 3 D - Lissajous Reconstruct object: acquire long enough to resolve neighboring sidebands k-space trajectory Phantom [http: //www. bic. mni. mcgill. ca/brainweb] VIEWS reconstruction 3 oscillators (10, 10. 3, 12. 6) k. Hz, T = 100 ms, G = 0. 07 T/m , Fo. V = 50 mm, 163 voxels 18

Frequency modulation in 3 D - Lissajous Phantom VIEWS reconstruction (10 k. Hz) VIEWS reconstruction (2. 5 k. Hz) 3 oscillators (2. 5, 2. 6, 3. 2) k. Hz, T = 100 ms, G = 0. 07 T/m , Fo. V = 50 mm, 413 voxels 19

VIEWS in k-space Phantom ART FFT 3 oscillators (5, 5. 2, 6. 3) k. Hz, T = 10 ms, G = 0. 07 T/m , Fo. V = 50 mm, 203 voxels Traditional MRI: 20

VIEWS in k-space ART reconstruction FFT k-space FFT reconstruction Inv. FFT https: //cds. ismrm. org/protected/19 MPresentations/eduvids/15/710 B/1 705/ 21

Beyond k-space? https: //cds. ismrm. org/protected/19 MPresentations/eduvids/15/710 B/1705/ Non-Traditional Encoding Methods: To k-Space & Beyond

VIEWS vs EVI • • High resolution, short scan times Arbitrary spatial inhomogeneities Single shot (single FID) Suppressed acoustics Engineered spatial resolution No phase-encoded directions Isotropic voxel bandwidth [Van der Zwaag et al. , “Improved echo volumar imaging (EVI) for functional MRI”, Mag. Res. Med. (2006)] 23

Thank you https: //i 3 m-detectors. com/mrilab Dr. J. Alonso Group leader M. Murbach Senior scientist Dr. A. Ríos CEO Prof. Dr. J. M. Benlloch Institute director Dr. F. Galve Senior scientist E. Pallás Mechanical engineer P. Borreguero Physicist Dr. J. P. Rigla Researcher Dr. E. Díaz Researcher Dr. J. M. Algarín Postdoc C. Gramage Electronics engineer D. Grau Ph. D student R. Bosch Electronics engineer J. M. González Ph. D Student

Non-linear-gradient people know how to reconstruct… 25

VIEWS at work 3 oscillators (2500 Hz, 2600 Hz, 3200 Hz), T = 120 ms, dt = 500 ns, G = 0. 07 T/m, 41 3 voxels, Fo. V = 50 mm, dx = 1. 2 mm [http: //www. bic. mni. mcgill. ca/brainweb] 26

Super-resolution Through-plane SR works fine In-plane SR controversial [Van Reeth et al. , “Super-resolution in magnetic resonance imaging: A review”, Conc. Mag. Res. A (2012)] [Tieng et al. , “MRI resolution enhancement: How useful are shifted images obtained by changing the demodulation frequency? ”, Mag. Res. Med (2010)] [Uecker et al. , “Reply to: MRI resolution enhancement: How useful are shifted images obtained by changing the demodulation frequency ? ”, Mag. Res. Med (2011)] 27