T2* quantification using multi-echo gradient echo sequences: a comparative study of different readout gradients

Seonyeong Shin, Seong Dae Yun, N. Jon ShahScientific 2023 · DOI

TL;DR

This study compares monopolar and bipolar readout gradients for T2 quantification in MRI, finding that bipolar gradients offer higher precision, especially in low SNR or short TE range scenarios. This makes bipolar gradients a more advantageous choice for accurate T2 mapping in challenging imaging conditions.

What Problem Does This Paper Address?

Magnetic Resonance Imaging (MRI) relies on various parameters to create detailed tissue contrast. Quantifying intrinsic tissue parameters, such as relaxation times like T2, allows for more effective interpretation of MR images and can serve as indices of disease severity. T2 quantification is commonly performed using multi-echo gradient echo (GRE) sequences. These sequences can employ either monopolar or bipolar readout gradients. Bipolar gradients offer a theoretical advantage by enabling shorter echo spacing times, which allows for acquiring more echoes within a given scan time. This increased number of echoes can lead to more precise T2 quantification. However, despite this potential, a direct comparative study investigating the performance of monopolar versus bipolar readout gradients for T2 quantification, especially under varying imaging conditions, was lacking. This paper aims to fill that gap by systematically comparing these two gradient types.

Methods

The researchers employed a multi-pronged approach:

  • Theoretical Investigation: They used the Cramér-Rao lower bound (CRLB) to theoretically predict the precision of T2* quantification for both monopolar and bipolar readout gradients. This provided a baseline expectation for their performance.
  • Computer Simulations: They created simulated datasets to evaluate the impact of various imaging parameters (flip angle, TR, TE range, and bandwidth) on T2* quantification accuracy and precision for both gradient types. This allowed for controlled experimentation without the complexities of real-world imaging.
  • Phantom Experiments: They used a specially designed phantom with known T2* values to experimentally validate the simulation findings. This step allowed them to assess performance in a controlled, but physically realized, environment.
  • In Vivo Experiments: Finally, they conducted experiments on healthy human volunteers to assess the performance of monopolar and bipolar gradients in a real-world clinical scenario.

Throughout these stages, they systematically varied parameters like flip angle, repetition time (TR), echo time (TE) range, acceleration factor (R), and receiver bandwidth (BW) to understand their influence on T2* quantification.

Datasets

The study utilized the following datasets:

  • Computer Simulation Datasets: Generated based on mathematical models of multi-echo GRE sequences, mimicking the properties of the phantom used in experiments.
  • Phantom Experiment Datasets: Acquired from a spherical phantom containing manganese (II) chloride solutions at varying concentrations to simulate different T2* values found in brain tissues.
  • In Vivo Experiment Datasets: Acquired from healthy human volunteers.

No new benchmark datasets were introduced; rather, existing experimental setups were used to generate data for comparison.

Key Insights and Ideas

  • Bipolar Gradients for Precision: The core finding is that bipolar readout gradients consistently provide higher precision (lower standard deviation) in T2* quantification compared to monopolar gradients, across simulations, phantom, and in vivo experiments.
  • SNR and TE Range Matter: The performance difference between monopolar and bipolar gradients is most pronounced in low Signal-to-Noise Ratio (SNR) or when using a shorter TE range. This is where bipolar gradients truly shine.
  • Time Efficiency: Bipolar gradients allow for shorter echo spacing times, enabling more echoes to be acquired in the same scan time. This directly contributes to improved precision.
  • Amplitude Modulation is a Factor: For bipolar gradients, amplitude modulation (AM) can introduce signal intensity variations that affect T2* quantification, particularly at shorter TE ranges. However, this effect can be effectively corrected.
  • Accuracy is Comparable with Sufficient Sampling: When the TE range is sufficiently long (i.e., comparable to or longer than the true T2*), both monopolar and bipolar gradients achieve similar accuracy. The advantage of bipolar gradients lies primarily in precision and efficiency.

Key Results

  • Higher Precision: In simulations, bipolar gradients showed a lower standard deviation of T2* values compared to monopolar gradients, a trend that held true in phantom and in vivo experiments. This difference was particularly notable at lower SNRs and shorter TE ranges.
  • Improved SNR with Bipolar: By reducing echo spacing, bipolar gradients can acquire more echoes, leading to better SNR and thus more robust T2* quantification, especially in challenging regions.
  • Anatomical Detail: In vivo T2* maps from bipolar readouts showed clearer anatomical structures, particularly in magnified views and under higher acceleration factors.
  • Amplitude Modulation Correction: While amplitude modulation was observed to affect bipolar readout T2* maps, correction methods effectively mitigated these distortions.

Limitations and Future Work

  • T2* Range Specificity: The study focused on T2 values specific to brain tissue at 3T. The authors suggest that the benefits of bipolar readouts might be even more pronounced in tissues with shorter T2 values (e.g., liver) where shorter TE ranges are often necessary.
  • Noise Estimation: The noise variance estimation in the study relied on background ROIs, which could be a limitation if noise fields are spatially varying.
  • Amplitude Modulation Correction: While effective, the need for amplitude modulation correction in bipolar readouts adds a step to the processing pipeline. Future work could focus on developing more integrated or automated correction methods.
  • Gradient Performance: The study acknowledges that practical implementations of ATE are constrained by gradient hardware capabilities (amplitude and slew rate). Further research could explore the impact of advanced gradient systems on bipolar readout performance.

Summary updated

Abstract

To quantify T2*, multiple echoes are typically acquired with a multi-echo gradient echo sequence using either monopolar or bipolar readout gradients. The use of bipolar readout gradients achieves a shorter echo spacing time, enabling the acquisition of a larger number of echoes in the same scan time. However, despite their relative time efficiency and the potential for more accurate quantification, a comparative investigation of these readout gradients has not yet been addressed. This work aims to compare the performance of monopolar and bipolar readout gradients for T2* quantification. The differences in readout gradients were theoretically investigated with a Cramér-Rao lower bound and validated with computer simulations with respect to the various imaging parameters (e.g., flip angle, TR, TE, TE range, and BW). The readout gradients were then compared at 3 T using phantom and in vivo experiments. The bipolar readout gradients provided higher precision than monopolar readout gradients in both computer simulations and experimental results. The difference between the two readout gradients increased for a lower SNR and smaller TE range, consistent with the prediction made using Cramér-Rao lower bound. The use of bipolar readout gradients is advantageous for regions or situations where a lower SNR is expected or a shorter acquisition time is required.
BibTeX
@inproceedings{shin2023t,
  author = {Seonyeong Shin and Seong Dae Yun and N. Jon Shah},
  title = {{T2* quantification using multi-echo gradient echo sequences: a comparative study of different readout gradients}},
  booktitle = {Scientific},
  year = {2023},
  doi = {10.1038/s41598-023-28265-0},
}
Figure