MENU

Click here to

×

Are you sure ?

Yes, do it No, cancel
  1. Home
  2. Meeting Program
  3. All Sessions
  4. All Multi-Disciplinary Scientific Session
  5. Evaluation of DIR-Based Lung Ventilation Imaging Against Hyperpolarized Gas Ventilation MRI and Hyperpolarized Gas Tagging MRI

Evaluation of DIR-Based Lung Ventilation Imaging Against Hyperpolarized Gas Ventilation MRI and Hyperpolarized Gas Tagging MRI

I Duarte1*, S Lam2 , T Cui3 , G Miller4 , W Garrison4 , J Mugler, III4 , M Shim4 , G Cates4 , F Yin1 , J Cai1,2 , (1) Duke University Medical Center, Durham, NC, (2) Hong Kong Polytechnic University, Hong Kong, (3) Rutgers Cancer Institute of New Jersey, New Brunswick, NJ, (4) University of Virginia, Charlottesville, VA

Presentations

(Thursday, 7/18/2019) 7:30 AM - 9:30 AM

Room: 225BCD

Purpose: To investigate deformable image registration (DIR)-based lung ventilation imaging technique by evaluating its correlation with hyperpolarized (HP) Helium-3 gas ventilation MRI reference images which provide a ground-truth measure of lung ventilation. Correlation between the reference ventilation images and ventilation maps computed from HP gas tagging MRIs, which provide ground-truth lung deformation, was also investigated.

Methods: Proton and HP Helium-3 gas tagging MRI images of the lungs for three healthy subjects were acquired at the end-of-inhalation (EOI) and end-of-exhalation (EOE) phases using a hybrid MRI technique during one same breath-hold. For each subject, a total of 5 image datasets were acquired in the single breath-hold maneuver: EOI/EOE pair of proton MRI (pMRI), EOI/EOE pair of HP gas tagging MRI (tMRI), and EOI HP gas ventilation (gVent). To measure the lung deformation from the tMRIs, the tagged grids’ center of mass was tracked from EOI to EOE, and the tagging-based displacement vector field (tDVF) was obtained within the lungs. The sparse tDVF, from approximately 400-500 uniformly distributed grids, was interpolated using a spline algorithm. The DIR-based displacement vector fields (dDVF) were extracted from the DIR of the EOI/EOE pMRI’s using the commercial DIR software Velocity, AI. Both, the tagging-based (tVent) and DIR-based (dVent) ventilation map surrogates were computed through the Jacobian Determinant of their respective DVFs. The HP gas MRI ventilation image was used as the reference to evaluate tVent and dVent, and voxel-wise Spearman correlation coefficients were calculated.

Results: The Spearman correlation coefficients between tVent and gVent were 0.47, 0.36, 0.18 for subjects 1, 2 and 3, respectively; with a SD of 0.15. The coefficients between dVent and gVent were 0.11, 0.07, 0.10 with a SD of 0.02.

Conclusion: Preliminary results show a moderate-to-low level of correlation for the tagging-based ventilation surrogates and a low level of correlation for DIR-based.

Funding Support, Disclosures, and Conflict of Interest: This work has been supported in part by the following NIH grants: R21CA195317, F31CA224980.

Keywords

Ventilation/perfusion, MRI, Deformation

Taxonomy

IM- MRI : Quantitative imaging/analysis

Contact Email

Additional Links