This repository contains a MATLAB script (script_writeGradientEcho3D_label_spoil_github_v0.m) used to generate a 3D Multi-Echo Gradient Echo (ME-GRE) MRI sequence using the open-source Pulseq framework.

This sequence is optimized for phase imaging, Quantitative Susceptibility Mapping (QSM), and parallel imaging reconstruction, featuring steady-state gradient/RF spoiling and built-in mapVBVD labels for GRAPPA reconstruction.

• Multi-Echo Acquisition: Acquires 5 echoes per TR using a unipolar flyback readout gradient design.
• GRAPPA Acceleration: Implements 1D phase-encode (Y-axis) undersampling with integrated Auto-Calibration Signal (ACS) lines.
• The sequence also allows for R=1 fully sampled acquisition, enabled by setting Ry = 1;
• Siemens mapVBVD Labels: Injects inline Pulseq labels (LIN, PAR, ECO, REF, IMA, NOISE) directly into the ADC blocks. This allows seamless data sorting and standard pipeline reconstruction using tools like mapVBVD.
• Spoiling & Steady State: * Continuous quadratic RF phase spoiling globally across all TRs (using an 84° increment for optimal steady-state stabilization).
  • Z-axis gradient phase-encode blips are perfectly rewound and combined with the spoiler (gzRewindAndSpoil) to maintain a constant net gradient moment per TR.
• Dummy Scans: Plays out 50 dummy TRs prior to acquisition with fully matched gradient structures to gracefully drive the magnetization into the steady state.
• Gx, Gy, and Gz gradients are flipped so that the online reconstructed image matches those reconstructed using the native Siemens GRE sequence.

By default, the script generates a protocol with the following parameters (which can be easily modified in the script):

• Matrix Size: 264 (RO) × 184 (PE) × 144 (PAR)
• FOV: 264 × 184 × 144 mm³
• Resolution: 1 mm isotropic
• TEs (Echo Times): 5.0, 11.0, 17.0, 23.0, 29.0 ms
• TR (Repetition Time): 35.0 ms
• Flip Angle: 15°
• Acceleration (Ry): 2
• ACS Lines: 32

To run this script, you will need:

1. MATLAB (R2019a or newer recommended)
2. Pulseq MATLAB Toolbox (v1.5.1): You must have the official Pulseq repository downloaded and added to your MATLAB path.
3. On the scanner console: from the resolution tab, set GRAPPA as the acceleration method using R=2 and 32 acs lines (integrated). From the sequence>special tab, select data handling -> ICE_STD to get the online recon to work.
4. Set Orientation to "Transversal" and phase encoding dir. to "R>>L". The slab position can also rotated/translated.
5. if fully sampled acquisition is desired set acceleration to "None", still using ICE_STD for data handling.

1. Clone this repository to your local machine.
2. Open script_writeGradientEcho3D_label_spoil_github.m in MATLAB.
3. Update the Pulseq directory path at the top of the script to point to your local installation:

   addpath('path/to/your/pulseq-1.5.1/matlab/')