Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

T1-PILOT: Optimized Trajectories for T1 Mapping Acceleration

Tamir Shor, Moti Freiman, Chaim Baskin, Alex Bronstein

TL;DR

Cardiac T1 mapping remains time-constrained and motion-sensitive. T1-PILOT presents a physics-informed, end-to-end framework that jointly optimizes non-Cartesian k-space trajectories and T1 decay estimation by embedding the T1 relaxation model into the acquisition and reconstruction process. A 3-stage optimization (reconstruction-guided pre-training, decay optimization, and per-sample refinement) enables high acceleration while preserving quantitative accuracy, demonstrated on the CMRxRecon dataset with superior PSNR and VIF compared to baselines. The results show that coupling the relaxation model with trajectory learning yields sharper, more accurate T1 maps at greater speeds, advancing practical rapid quantitative MRI for the heart.

Abstract

Cardiac T1 mapping provides critical quantitative insights into myocardial tissue composition, enabling the assessment of pathologies such as fibrosis, inflammation, and edema. However, the inherently dynamic nature of the heart imposes strict limits on acquisition times, making high-resolution T1 mapping a persistent challenge. Compressed sensing (CS) approaches have reduced scan durations by undersampling k-space and reconstructing images from partial data, and recent studies show that jointly optimizing the undersampling patterns with the reconstruction network can substantially improve performance. Still, most current T1 mapping pipelines rely on static, hand-crafted masks that do not exploit the full acceleration and accuracy potential. In this work, we introduce T1-PILOT: an end-to-end method that explicitly incorporates the T1 signal relaxation model into the sampling-reconstruction framework to guide the learning of non-Cartesian trajectories, crossframe alignment, and T1 decay estimation. Through extensive experiments on the CMRxRecon dataset, T1-PILOT significantly outperforms several baseline strategies (including learned single-mask and fixed radial or golden-angle sampling schemes), achieving higher T1 map fidelity at greater acceleration factors. In particular, we observe consistent gains in PSNR and VIF relative to existing methods, along with marked improvements in delineating finer myocardial structures. Our results highlight that optimizing sampling trajectories in tandem with the physical relaxation model leads to both enhanced quantitative accuracy and reduced acquisition times.

T1-PILOT: Optimized Trajectories for T1 Mapping Acceleration

TL;DR

Cardiac T1 mapping remains time-constrained and motion-sensitive. T1-PILOT presents a physics-informed, end-to-end framework that jointly optimizes non-Cartesian k-space trajectories and T1 decay estimation by embedding the T1 relaxation model into the acquisition and reconstruction process. A 3-stage optimization (reconstruction-guided pre-training, decay optimization, and per-sample refinement) enables high acceleration while preserving quantitative accuracy, demonstrated on the CMRxRecon dataset with superior PSNR and VIF compared to baselines. The results show that coupling the relaxation model with trajectory learning yields sharper, more accurate T1 maps at greater speeds, advancing practical rapid quantitative MRI for the heart.

Abstract

Cardiac T1 mapping provides critical quantitative insights into myocardial tissue composition, enabling the assessment of pathologies such as fibrosis, inflammation, and edema. However, the inherently dynamic nature of the heart imposes strict limits on acquisition times, making high-resolution T1 mapping a persistent challenge. Compressed sensing (CS) approaches have reduced scan durations by undersampling k-space and reconstructing images from partial data, and recent studies show that jointly optimizing the undersampling patterns with the reconstruction network can substantially improve performance. Still, most current T1 mapping pipelines rely on static, hand-crafted masks that do not exploit the full acceleration and accuracy potential. In this work, we introduce T1-PILOT: an end-to-end method that explicitly incorporates the T1 signal relaxation model into the sampling-reconstruction framework to guide the learning of non-Cartesian trajectories, crossframe alignment, and T1 decay estimation. Through extensive experiments on the CMRxRecon dataset, T1-PILOT significantly outperforms several baseline strategies (including learned single-mask and fixed radial or golden-angle sampling schemes), achieving higher T1 map fidelity at greater acceleration factors. In particular, we observe consistent gains in PSNR and VIF relative to existing methods, along with marked improvements in delineating finer myocardial structures. Our results highlight that optimizing sampling trajectories in tandem with the physical relaxation model leads to both enhanced quantitative accuracy and reduced acquisition times.

Paper Structure

This paper contains 13 sections, 4 equations, 2 figures, 1 table.

Table of Contents

  1. Introduction
  2. Method
  3. Problem Definition
  4. Optimization
  5. Reconstruction-Guided Pre-Training
  6. Decay Optimization
  7. Per-Sample Reconstruction & Decay Refinement
  8. Results
  9. Data
  10. Experimental Setup
  11. Quantitative Results
  12. Qualitative Results
  13. Conclusion

Figures (2)

  • Figure 1: T1-PILOT Pipeline - blue arrows denote the forward process, and orange arrows denote backpropagation.
  • Figure 2: Map Estimations across methods - Full indicates map estimation without undersampling. For each baseline, we specify ROI PSNR (left) and VIF (right) compared to the fully-sampled map estimation. Our method's relative advantage over baselines is highlighted in red squares.