WP-005 - 5D Flow MRI using Pilot Tone for Cardiac and Respiratory Self-Gating

Wednesday, February 12, 2020

5:00 PM – 7:00 PM

Location: Canary 3 & 4 Foyer

Background: Current clinical methods for 4D flow MRI rely on ECG gating and respiratory navigators, leading to prolonged and unpredictable scan times. Recent advances have enabled cardiac and respiratory motion-resolved 5D flow with a fixed scan time, using a free-running framework^{[1, 2]}. This protocol employed a 3D radial phyllotaxis trajectory^[3], and periodic readouts along the superior-inferior direction allow for both cardiac and respiratory self-gating^[1], eliminating the need for an ECG signal and navigator scans. Alternatively, Pilot Tone navigation allows for the extraction of these physiologic signals in parallel to the acquisition^[4,5,6]. In this study, we explore the initial feasibility of using Pilot Tone for ECG- and navigator-free 5D flow MRI.

Methods: Six volunteers (4F, age: 28.0±4.8 years) were scanned on a 1.5T MAGNETOM Sola (Siemens Healthcare, Erlangen, Germany), using a 12-channel body coil with an integrated Pilot Tone transmitter. Each volunteer was scanned using a prototype radial 5D Flow MRI sequence covering the whole heart and 4D Flow sequence covering the aorta with the parameters listed in Fig. 2. Cardiac and respiratory gating signals were retrospectively extracted from the multi-channel Pilot Tone signals as described by Fig. 1 and subsequently used to bin the data into 4 respiratory and 17-21 cardiac frames, depending on the subject’s heartrate. Cardiac binning accuracy was quantified by the standard deviation of the difference between Pilot Tone and ECG triggers. The resulting 5D flow datasets were reconstructed using XD-GRASP^[7]. Flow curves were calculated for both the 4D and 5D Flow data sets in the ascending aorta, descending aorta and aortic arch using the Siemens 4D Flow v2.4 Software. The net flow volume and the peak flow were compared using the Wilcoxon signed rank test.

Results: The acquisition time was 9min 01s for 5D Flow and 11min 19s ± 4min 24s for 4D Flow. Pilot Tone cardiac triggers matched the ECG triggers with an average error of 4.8±3.4% Aortic flow curves (Fig. 2) derived from 5D flow underestimated peak systolic flow by 23±11%. Net flow and peak flow values (Fig. 3) showed that 5D Flow tends to underestimate the flow measurements by 11±14% and 22±11% respectively (p < 0.05 in 4 out of 6 measurements).

Conclusion: ECG- and navigator-free cardiac and respiratory motion-resolved 5D Flow with a fixed scan time is feasible using Pilot Tone, providing a means for whole-heart evaluation of hemodynamics with shorter scan time than conventional 4D Flow. The tendency to underestimate flow values in 5D Flow has been reported before^[2], and may be caused by temporal undersampling and regularization. While further optimization of acquisition and reconstruction parameters is needed to improve the consistency between 5D flow and the 4D flow reference standard, the use of Pilot Tone gating, which is independent of the acquisition, provides a new method for fully self-gated 5D flow.

Presenting Author(s)

Mariana Falcão, MSc

PhD Student
University Hospital (CHUV) and University of Lausanne (UNIL)

Disclosure: Disclosure information not submitted.

Primary Author(s)

Mariana Falcão, MSc

PhD Student
University Hospital (CHUV) and University of Lausanne (UNIL)

Disclosure: Disclosure information not submitted.

Co-Author(s)

Liliana Ma

MD/PhD student
Northwestern University

Disclosure: Disclosure information not submitted.
Lorenzo Di Sopra, DO, MSc

PhD student
University Hospital (CHUV) and University of Lausanne (UNIL)

Disclosure: Disclosure information not submitted.
MB

Mario Bacher, MSc

PhD Student
Siemens Healthcare GmbH and University Hospital (CHUV) and University of Lausanne (UNIL)

Disclosure: Disclosure information not submitted.
Davide Piccini, PhD

Senior Scientist
Siemens Healthcare Switzerland

Disclosure: Disclosure information not submitted.
JY

Jérôme Yerly, PhD

University Hospital (CHUV) and University of Lausanne (UNIL)

Disclosure: Disclosure information not submitted.
PS

Peter Speier, PhD

Principal Key Expert
Siemens Healthcare GmbH

Disclosure: Siemens Healthcare GmbH (Individual(s) Involved: Self, Products/Services: MR products): Other Financial or Material Support, Salary, patent royalties, Stockholder/Ownership Interest (excluding diversified mutual funds)
Michael Markl, PhD

Professor & Vice Chair for Research
Northwestern University

Disclosure: Circle Cardiovascular Imaging: Consulting Fees (e.g., advisory boards), Grant/Research Support; Cryolife Inc (Individual(s) Involved: Self, Products/Services: grant): Grant/Research Support; NXT Biomedical (Individual(s) Involved: Self, Products/Services: Consultant): Consulting Fees (e.g., advisory boards); Siemens Healthcare: Grant/Research Support
Matthias Stuber, PhD

Professor
University Hospital (CHUV) and University of Lausanne (UNIL)

Disclosure: Disclosure information not submitted.
CR

Christopher W. Roy, PhD, MSc

Postdoctoral Researcher
University Hospital (CHUV) and University of Lausanne (UNIL)

Disclosure: Disclosure information not submitted.