Artifacts-reduced 2D cine SSFP with flow compensation for myocardial BOLD imaging

Xiangzhi Zhou, Sotirios A. Tsaftaris, Ying Liu, Richard Tang, Rachel Klein, Sven Zuehlsdorff, Debiao Li, Rohan DharmakumarJournal of Cardiovascular Magnetic Resonance 2010 · DOI

Abstract

Robust image quality is critical for reliable detection and evaluation of myocardial oxygenation changes with blood-oxygen-level-dependent (BOLD) imaging. Recently, balanced SSFP methods have been employed to overcome image quality limitations associated with myocardial BOLD methods. However, the long TRs required for BOLD contrast can lead to unwanted flow/motion artifacts, ultimately compromising image quality.

Introduction

Myocardial BOLD imaging is a non-invasive technique used to detect changes in heart muscle oxygenation, which is essential for identifying coronary artery disease. Balanced Steady-State Free Precession (SSFP) sequences are often used for this purpose due to their high signal-to-noise ratio. However, a significant challenge exists: to achieve high BOLD sensitivity, a relatively long Repetition Time (TR) is required. This long TR increases the sequence's sensitivity to cardiac motion and blood flow, often resulting in "ghosting" artifacts and signal inhomogeneity that can obscure diagnostic information. This study proposes a method to maintain long-TR contrast while eliminating these artifacts using first-order flow and motion compensation.

Methods

The researchers conducted a study using a canine model (n=6) in a 1.5 Tesla scanner. They compared three distinct imaging approaches:

  1. Approach A (Conventional): A short TR of 3.5 ms, serving as the gold standard for image stability but lacking BOLD sensitivity.
  2. Approach B (Uncompensated): A long TR of 6.2 ms without motion correction, providing BOLD contrast but prone to artifacts.
  3. Approach C (Flow-Compensated): A long TR of 6.2 ms using first-order flow/motion compensation.

The "first-order compensation" utilizes specifically designed magnetic gradient waveforms to null the phase shifts caused by constant velocity motion, such as blood flow. Experts scored the resulting images for Ghost Artifacts (GA) and Myocardial Inhomogeneity (MI) on a scale of 1 (best) to 5 (worst). The method was also validated in a coronary stenosis model to evaluate its ability to detect oxygenation deficits under stress.

Key Results

The study demonstrated that the trade-off between BOLD sensitivity and image quality is not mandatory. Flow compensation effectively "reset" the image quality of the long-TR sequence to match that of the short-TR conventional sequence.

Artifact scores for the basal slice show that flow compensation (Approach C) significantly reduces ghosting (GA) and inhomogeneity (MI) compared to the uncompensated long-TR sequence (Approach B), achieving parity with the short-TR standard (Approach A).

Statistical analysis showed no significant difference in artifact scores between Approach A and Approach C. In contrast, Approach B performed significantly worse (p < 0.01), particularly in the basal sections of the heart where motion and flow patterns are most complex.

Clinical Utility and Visualization

By reducing artifacts, the researchers were able to clearly visualize signal loss in specific myocardial territories (e.g., the LCX territory) during both the contraction (systole) and relaxation (diastole) phases of the heartbeat. In stenosis models, flow-compensated images allowed for the clear identification of oxygenation deficits that were previously obscured.

Cardiac MRI images (LS: Late Systole, LD: Late Diastole) demonstrating how flow compensation allows for the clear visualization of myocardial territories and potential signal loss (arrows) throughout the cardiac cycle.

Conclusion and Future Work

First-order flow compensation successfully mitigates the artifacts associated with long-TR balanced SSFP BOLD imaging. While this study utilized a small animal model and focused on constant velocity motion, the results provide a strong basis for clinical translation. Future work will involve evaluating this strategy in human patients with suspected coronary artery disease and exploring higher-order compensation for more complex motion environments.

Summary updated

Abstract

Robust image quality is critical for reliable detection and evaluation of myocardial oxygenation changes with blood-oxygen-level-dependent (BOLD) imaging. Recently, balanced SSFP methods have been employed to overcome image quality limitations associated with myocardial BOLD methods. However, the long TRs required for BOLD contrast, can lead to unwanted flow/motion artifacts, ultimately compromising image quality.
BibTeX
@article{zhou2010artifacts,
  author = {Xiangzhi Zhou and Sotirios A. Tsaftaris and Ying Liu and Richard Tang and Rachel Klein and Sven Zuehlsdorff and Debiao Li and Rohan Dharmakumar},
  title = {{Artifacts-reduced 2D cine SSFP with flow compensation for myocardial BOLD imaging}},
  journal = {Journal of Cardiovascular Magnetic Resonance},
  year = {2010},
  doi = {10.1186/1532-429x-12-s1-p68},
}
Figure