Initial Feasibility of Free-Breathing Whole-Heart Isotropic 3D Diffusion Weighted Imaging in Humans using a Cone Trajectory

Presenting Author(s)

• 

  Zachary Player, BA, BSc

  PhD Student
  Case Western Reserve University, Cleveland Clinic
  Cleveland Heights, Ohio, United States

Primary Author(s)

• 

  Zachary Player, BA, BSc

  PhD Student
  Case Western Reserve University, Cleveland Clinic
  Cleveland Heights, Ohio, United States

Co-Author(s)

• 

  Moses Cook, PhD

  Postdoctoral Fellow
  Cleveland Clinic
  Cleveland, Ohio, United States
• DM

  Dingheng Mai

  PhD Candidate
  Cleveland Clinic
  Cleveland, Ohio, United States
• TR

  Tassia Ribeiro Salles Moura, MSc

  Graduate Researcher
  Cleveland Clinic / Cleveland State University
  Westlake, Ohio, United States
• PS

  Pasquale Santangeli, MD, PhD

  Director
  Ventricular Tachycardia (VT) program in the Cardiac Electrophysiology and Pacing Section of the Tomsich Family Department of Cardiovascular Medicine, Miller Family Heart, Vascular & Thoracic Institute, Cleveland Clinic
  Cleveland, Ohio, United States
• 

  Animesh Tandon, MD, MS

  Director of Cardiovascular Innovation
  Cleveland Clinic Children's
  Cleveland, Ohio, United States
• OW

  Oussama Wazni, MD, MBA

  Physician
  Cardiac Electrophysiology and Pacing at Sydell and Arnold Miller Family Heart, Vascular & Thoracic Institute, Cleveland Clinic
  Cleveland, Ohio, United States
• SJ

  Stephen Jones, MD, PhD

  Staff Physician
  Cleveland Clinic
  Cleveland, Ohio, United States
• 

  Danielle Kara, PhD

  Staff Scientist
  Cleveland Clinic
  Cleveland, Ohio, United States
• YL

  Yuchi Liu, PhD

  Research Scientist
  Siemens Medical Solutions USA, Inc.
  Cleveland, Ohio, United States
• 

  Christopher Nguyen, PhD, FSCMR

  Director, Cardiovascular Innovation Research Center
  Cleveland Clinic
  Cleveland, Ohio, United States

Background: Cardiac diffusion weighted imaging (DWI) has been under development for more than a decade [1] and has seen significant progress during the period. However, it remains limited to 2D, with typical spatial resolution of 2.7x2.7x8 mm³. A common acquisition approach uses echo planar imaging (EPI), which is susceptible to cardiac distortion. Although 2D cardiac diffusion has many potential applications, it remains challenging to interpret diffusion biomarkers given large slice thicknesses.

An alternative acquisition method to EPI is the 3D cone readout[2]. This trajectory is signal-to-noise ratio efficient and resistant to distortion. The cone trajectory is natively 3D isotropic, enhancing its effectiveness for cardiac DWI. Here, we describe a method of whole-heart 3D isotropic DWI in the human heart using an interleaved 3D cone trajectory with second-order motion compensated (M2) diffusion encoding gradients and spin-echo acquisition. We hypothesized that the apparent diffusion coefficient (ADC) of 3D DWI with cone readout will be qualitatively similar to mean diffusivity (MD) in a reference DWI with EPI readout.

Methods: We employ an Archimedean-spiral-based 3D cone trajectory [2] optimized with SigPy with 8 interleaves and M2 diffusion encoding [1] (Figure 1) on a 3T MRI scanner (MAGNETOM Cima.X, Siemens Healthineers, Forchheim, Germany). 3 healthy volunteers were scanned under an IRB approved protocol. We used the following sequence parameters: TR/TE=1heartbeat/40.45 ms, resolution = 2.5x2.5x2.5mm³, readout length = 6.07 ms, b-value = 500s/mm², directions = 3. To acquire the whole heart, the sequence was 1528 heartbeats long, corresponding to a 21-31 minute acquisition per DWI. For comparison, we also acquired DWI using a reference 2D cardiac M2 spin-echo EPI diffusion research sequence [3]. We reconstructed the images with SigPy using NUFFT and SENSE reconstruction, and region-optimized virtual coils (ROVir) [4] and calculated ADC.

Results: The 3D DWI with cones readout yielded images with good b50 and b500 DWI fidelity, including less prevalent distortion compared to the EPI reference, more similar to the bSSFP CINE (Figure 2). Moreover, we found similar ADC with 3D cones compared to MD of 2D EPI readout in this limited cohort after a paired t-test (p=0.91) (Figure 3), with a mean ADC of 1.543 ± 130 x 10⁻³ mm²/s and a mean MD of 1.553 ± 65 x 10⁻³ mm²/s.

Conclusion: We demonstrate feasibility of whole heart 3D isotropic cardiac diffusion-weighted MRI. Future work will aim to more robustly compare results between the proposed sequence and standard diffusion sequences in a larger population. Furthermore, we plan to optimize the protocol to eventually fit more sophisticated diffusion models to enable whole heart disease characterization.

A. Pulse sequence diagram for spin echo 3D diffusion cone with M2 diffusion encoding. B. The sphere in k-space described by individual cones of the acquisition. C. The trajectory of an individual cone. D. Volume renders of heart captured with our sequence. E. Transverse, coronal, and sagittal views of a single 3D DWI with b=500 acquired with our cone sequence.
A. Short axis view acquired with 2D diffusion EPI sequence. B. Short axis view acquired with our 3D cone sequence. C. Short axis view acquired with CINE sequence.
A-C. Apical to basal slices acquired with 2D diffusion EPI from Volunteer 1. D-F. Matched slices acquired from same volunteer with our 3D diffusion cone sequence. H-J. Apical to basal slices acquired with 2D diffusion EPI from Volunteer 2. K-M. Matched slices acquired from same volunteer with our 3D diffusion cone sequence.