8000 Image Reconstruction · imr-framework/mrf Wiki · GitHub
[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Skip to content

Image Reconstruction

Enlin Qian edited this page Sep 4, 2019 · 4 revisions

This script reconstructs images based on raw data and trajectory file.

Method

To run MRF_recon.m, MIRT (Michigan Image Reconstruction Toolbox) is required. The script will ask the user to locate a .dat raw data file and a .mat trajectory file, convert the .dat file to .mat file, multiply the density compensation function (dcf) with raw data, and then reconstruct the data using sliding window and complex coil combination method. The following details each step.

  1. The script reads a .dat raw data and convert it to .mat raw data as shown below, then concatenates the data from different partitions. A sample raw data file can be found in Sample_Data/T1_plane_raw_data.dat. The data is acquired using the sequence described in Sequence design part.

2. The script reads a .mat trajectory file as shown blow. A sample trajectory file can be found in **Sample_Data/K_Traj4rec.mat**. The algorithm compares the length of raw data file and trajectory file to see if they match. If the raw data is longer than the trajectory file, the rewinder in the end will be cut.

3. In this implementation, because the spiral has 48 shots, the sliding window is 24 points before and 24 points after for each time points. The algorithm combines all of them to complete K-space. The idea can be shown below.

4. Depending on the location of the coil, when combining data from different coils, their phase needs to be corrected. This idea is well explained in [1].

Then it will generate reconstructed image as shown below (all 1000 time points).

Demo

To test this module, please run Image_Reconstruction/MRF_recon_demo.m. This script calls MRF_recon.m and reconstruct images based on input trajectory and raw data file.

Reference

[1] Roemer P, Edelstein W, Hayes C, Souza S, Mueller O. The NMR phased array. Magn Reson Med. 1990;16(2):192-225. doi:10.1002/mrm.1910160203

Clone this wiki locally
0