Cardiac phase-resolved late gadolinium enhancement imaging

Sebastian Weingärtner, Ömer Burak Demirel, Francisco Gama, Iain Pierce, Thomas A. Treibel, Jeanette Schulz‐Menger, Mehmet AkçakayaFrontiers in Cardiovascular Medicine 2022 · DOI

TL;DR

This paper introduces a novel method for creating "functional" Late Gadolinium Enhancement (LGE) cardiac MRI images that capture scar tissue throughout the entire cardiac cycle, not just a single snapshot. This allows for the visualization of scar motility and offers a more comprehensive assessment of myocardial viability, potentially improving diagnostic accuracy.

What Problem Does This Paper Address?

Cardiac MRI's gold standard for detecting myocardial scar, Late Gadolinium Enhancement (LGE), is typically performed at a single point in diastole. While effective for identifying scar, this approach misses the dynamic nature of the heart. It doesn't allow for the assessment of scar movement (motility) or direct comparison of scar appearance across different cardiac phases. This limitation necessitates separate scans (like CINE MRI for motion) and requires clinicians to mentally fuse information from different acquisitions, which can be cumbersome and potentially lead to missed insights. The goal here is to integrate functional cardiac motion information directly into the LGE assessment.

Methods

The authors propose a three-step approach to achieve phase-resolved LGE imaging:

  1. Acquisition of Phase-Resolved Imaging Data: They extend their previously developed TOPAZ technique to acquire multiple cardiac phase-resolved images with varying T1 weighting. This is achieved using a Look-Locker sequence that samples the magnetization recovery curve at multiple points within each cardiac phase. Crucially, this acquisition is prospectively triggered to the ECG, and dummy pulses are used to maintain a consistent steady-state recovery, even with variable heart rates.
  2. Generation of Semi-Quantitative T1 Maps: For each cardiac phase, the acquired T1-weighted images are used to generate semi-quantitative T1 maps. A simplified two-parameter fit model is employed, assuming complete magnetization inversion, to estimate apparent T1 times.
  3. Synthesization of LGE Contrast: Using the T1 maps, LGE contrast is retrospectively synthesized for each cardiac phase. A specific "synthetic inversion time" (Tsyn) is chosen to null the signal from healthy myocardium, creating the characteristic hyper-enhancement of scar tissue against a dark background. This synthetic inversion time is applied consistently across all cardiac phases, ensuring comparable LGE contrast for cross-phase comparison.

Datasets

The proposed method was evaluated using:

  • Phantom Experiments: Two sets of phantom experiments were conducted on a 3T scanner. One used spheres mimicking blood pool and healthy myocardium to assess contrast-to-noise ratio (CNR) and contrast homogeneity throughout the cardiac cycle. Another used a bottle phantom to evaluate performance at different simulated heart rates.
  • In-Vivo Experiments:
    • Healthy Subjects: Six healthy subjects were scanned at 3T with 40 ms temporal resolution.
  • Patients: Twenty patients were scanned at 1.5T with 80 ms temporal resolution.
  • Additional Patients: Four patients were scanned at 3T with 60 ms temporal resolution.

The paper does not introduce new benchmark datasets but rather evaluates the method on existing patient and healthy subject data, comparing it to conventional LGE imaging.

Key Insights and Ideas

  • Bridging the Gap: The core innovation is the ability to combine the functional information from CINE MRI (motion) with the tissue characterization of LGE (scar) into a single, phase-resolved acquisition. This eliminates the need for separate scans and mental fusion of data.
  • Scar Motility Visualization: The ability to see scar tissue throughout the cardiac cycle opens up the possibility of visualizing scar motility. This could be particularly useful in cases where scar is difficult to discern from surrounding tissue or blood pool.
  • Consistent Contrast is Key: The method's success hinges on generating consistent LGE contrast across all cardiac phases by retrospectively synthesizing the LGE effect using T1 maps. This ensures that comparisons between phases are valid.
  • Trade-offs in Clinical Application: While promising, the clinical application at 1.5T and even at 3T in patients revealed lower image resolution and higher noise resilience compared to conventional LGE. This highlights the ongoing challenge of balancing temporal resolution with image quality in dynamic cardiac imaging.

Key Results

  • Phantom Performance: The phantom experiments demonstrated consistent contrast throughout the cardiac cycle with a low coefficient of variation (CoV) of 2.55 ± 0.42% for CNR.
  • Healthy Subjects (3T): Reliable LGE contrast was achieved with thorough suppression of healthy myocardium. Moderate variation in blood-myocardium contrast was observed (CoV of 18.2 ± 3.51%). Temporal resolution of 40 ms was achieved.
  • Patients (1.5T): Images showed scar tissue throughout the cardiac cycle, but with noticeably lower imaging resolution and noise resilience compared to the reference technique. Temporal resolution of 80 ms was achieved.
  • Patients (3T): Scar tissue was visualized throughout the cardiac cycle, with some phases showing higher noise variability. The reference scan provided superior resolution and noise resilience.

Limitations and Future Work

  • Noise and Resolution in Patients: The most significant limitation is the observed higher noise levels and reduced imaging resolution in patient cohorts compared to conventional LGE, particularly at 1.5T. This is attributed to later acquisition times after contrast injection, leading to longer post-contrast T1 times and increased noise susceptibility.
  • Temporal Resolution vs. Noise: The paper acknowledges the need to improve the trade-off between spatiotemporal resolution and noise resilience to match reference LGE image quality.
  • Future Directions:
    • Regularized Reconstruction: Integrating regularized reconstruction schemes (similar to those used for TOPAZ) could enhance noise resilience and spatiotemporal resolution.
  • Deep Learning for Automation: Utilizing deep learning for automatic identification of the remote myocardium would enable automatic selection of the synthetic inversion time, streamlining the process.
  • Functional Black-Blood LGE: Extending the method to generate functional black-blood LGE images by combining T1 and T2 quantification could improve sub-endocardial scar detection.
  • Whole-Heart Coverage: Implementing simultaneous multi-slice (SMS) imaging could enable whole-heart coverage in fewer breath-holds, facilitating clinical translation.
  • Cardiac Function Quantification: Further work is needed to tailor the approach for functional volumetry, potentially by employing retrospective ECG gating or exploring free-breathing acquisition methods.

Summary updated

Abstract

Late gadolinium enhancement (LGE) with cardiac magnetic resonance (CMR) imaging is the clinical reference for assessment of myocardial scar and focal fibrosis. However, current LGE techniques are confined to imaging of a single cardiac phase, which hampers assessment of scar motility and does not allow cross-comparison between multiple phases. In this work, we investigate a three step approach to obtain cardiac phase-resolved LGE images: (1) Acquisition of cardiac phase-resolved imaging data with varying T 1 weighting. (2) Generation of semi-quantitative <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" id="M1"><mml:msubsup><mml:mrow><mml:mstyle class="text"><mml:mtext mathvariant="bold">T</mml:mtext></mml:mstyle></mml:mrow><mml:mrow><mml:mstyle class="text"><mml:mtext mathvariant="bold">1</mml:mtext></mml:mstyle></mml:mrow><mml:mrow><mml:mstyle class="text"><mml:mtext mathvariant="bold">*</mml:mtext></mml:mstyle></mml:mrow></mml:msubsup></mml:math> maps for each cardiac phase. (3) Synthetization of LGE contrast to obtain functional LGE images. The proposed method is evaluated in phantom imaging, six healthy subjects at 3T and 20 patients at 1.5T. Phantom imaging at 3T demonstrates consistent contrast throughout the cardiac cycle with a coefficient of variation of 2.55 ± 0.42%. In-vivo results show reliable LGE contrast with thorough suppression of the myocardial tissue is healthy subjects. The contrast between blood and myocardium showed moderate variation throughout the cardiac cycle in healthy subjects (coefficient of variation 18.2 ± 3.51%). Images were acquired at 40–60 ms and 80 ms temporal resolution, at 3T and 1.5, respectively. Functional LGE images acquired in patients with myocardial scar visualized scar tissue throughout the cardiac cycle, albeit at noticeably lower imaging resolution and noise resilience than the reference technique. The proposed technique bears the promise of integrating the advantages of phase-resolved CMR with LGE imaging, but further improvements in the acquisition quality are warranted for clinical use.
BibTeX
@inproceedings{weingartner2022cardiac,
  author = {Sebastian Weingärtner and Ömer Burak Demirel and Francisco Gama and Iain Pierce and Thomas A. Treibel and Jeanette Schulz‐Menger and Mehmet Akçakaya},
  title = {{Cardiac phase-resolved late gadolinium enhancement imaging}},
  booktitle = {Frontiers in Cardiovascular Medicine},
  year = {2022},
  doi = {10.3389/fcvm.2022.917180},
}
Figure