Interventional MRI - Methods

1354902 - Real-time Automatic Catheter Balloon Tracking During MR-guided Cardiac Catheterization

Friday, January 27, 2023

8:50 AM – 9:00 AM

Location: Silver Pearl 2

Presenting Author(s)

RV

Rohini Vidya Shankar, PhD

Research Associate
King's College London
London, United Kingdom

Primary Author(s)

RV

Rohini Vidya Shankar, PhD

Research Associate
King's College London
London, United Kingdom

Co-Author(s)

LH

Li Huang, PhD

Research Associate
King's College London
London, England, United Kingdom
RN

Radhouene Neji, PhD

Siemens Research Scientist
King's College London, United Kingdom
GK

Grzegorz T. Kowalik, PhD

Research Associate
King's College London
London, England, United Kingdom
AN

Alexander P. Neofytou, MSc

PhD Student
King's College London
London, England, United Kingdom
RM

Ronald Mooiweer, PhD

MRI Scientist & Research Associate
Siemens Healthcare & King's College London, United Kingdom
TM

Tracy Moon

Senior Paediatric MRI Radiographer
Guy's and St Thomas' NHS Foundation Trust, United Kingdom
NM

Nina Mellor

Senior Paediatric MRI Radiographer
Guy's and St Thomas' NHS Foundation Trust, United Kingdom
Reza Razavi, MD

Professor of Paediatric Cardiovascular Science
King's College London
London, England, United Kingdom
Kuberan Pushparajah, MD

Clinical Senior Lecturer in Paediatric Cardiology
King's College London
London, England, United Kingdom
SR

Sébastien Roujol, PhD

Reader in Medical Imaging
King's College London
London, England, United Kingdom

Disclosure(s):

Kuberan Pushparajah, MD: No financial relationships to disclose

Sébastien Roujol, PhD: No financial relationships to disclose

Background: MR-guided cardiac catheterization is an attractive alternative to X-ray guidance due to the lack of ionizing radiation and high soft tissue contrast.^1–3 Current passive tracking approaches require frequent manual manipulation of the imaging plane to follow the gadolinium-filled balloon during catheter navigation. In this study, we present the initial evaluation in patients of a novel CMR sequence that enables automatic real-time tracking and visualization of the balloon during catheter navigation.

Methods: The proposed approach⁴ has two imaging modes: Calibration and Runtime. For improved contrast between the balloon and anatomy, partial saturation (pSAT)⁵ and fat suppression pulses are used. First, in Calibration mode, a fixed stack of contiguous coronal slices (n=10–20, 10 mm thickness, pSAT=90°) is acquired in < 2–3s. Real-time image-processing (intensity thresholding and pattern matching within the prescribed shim box) is performed to identify the initial 3D location of the balloon. The sequence then automatically switches to Runtime mode where 3 parallel slices (10 mm thickness, pSAT=30–50°) are acquired continuously in the orientation of interest for optimal visualization of the procedure. These 3 slices are automatically adjusted to intersect the balloon in the central slice based on the 3D coordinates obtained from Calibration. During Runtime, the balloon coordinates are continuously estimated from real-time image-processing of the 3 slices (intensity thresholding, pattern matching, and spatiotemporal constraints). If the balloon is detected in one of the outer slices, the 3 slices are shifted to maintain the catheter in the central slice. The sequence switches back to Calibration if the balloon is lost for >5 dynamics.
The proposed framework was evaluated in a 3D printed heart phantom and in 2 patients undergoing MR-guided cardiac catheterization. All imaging was performed on a 1.5T MRI scanner (MAGNETOM Aera, Siemens Healthcare, Erlangen, Germany). The sequence parameters were: 2D single-shot bSSFP acquisition (TR/TE=2.44/1.02 ms, FA=50°, FOV = 450×450 mm², resolution=1.4×1.4 mm², slice thickness=10 mm, bandwidth=1010 Hz/px, GRAPPA factor=2, partial Fourier=5/8).

Results: Figure 1 shows still frames of the phantom with the balloon successfully detected during Calibration and tracked during Runtime. Figures 2&3 show the application in 2 patients, where the balloon was initially detected in the Calibration stack and then tracked with slice repositioning during Runtime while the catheter was manipulated. During Calibration, the balloon was automatically identified with 100% accuracy both in phantom and in vivo. During Runtime, the detection accuracy was 97.4% (85.7% in the central slice) in the phantom and 77.6±23.8% (59.1±2.8% in the central slice) in vivo.

Conclusion: The proposed approach enables continuous and robust automated tracking of the catheter and imaging slices during MR-guided cardiac catheterization.