Focus on MR Optimisation - Bandwidth

แชร์
ฝัง
  • เผยแพร่เมื่อ 3 ธ.ค. 2024

ความคิดเห็น • 16

  • @Lanier8
    @Lanier8 3 หลายเดือนก่อน

    AWESOME VIDEO! THANKS FOR THE HELP. Much appreciated.

    • @everything-mri
      @everything-mri  3 หลายเดือนก่อน

      My pleasure. Thank you so much for watching 🙏🏻

  • @anniesshenanigans3815
    @anniesshenanigans3815 11 หลายเดือนก่อน +1

    thanks for the info!! I was just talking to a coworker a few days ago about bandwidth and she brought up questions that I could not answer. I have a good grasp on when to use higher bandwidth (along with other adjustments) to get better images, but explaining it isn't as easy.

    • @everything-mri
      @everything-mri  11 หลายเดือนก่อน

      Thank you so much for watching. The bandwidth can be one of those parameters that we do not manipulate that frequently as we are predominantly concerned about potential implications. Sometimes, as you correctly said, simply there seem to be a lot of unanswered questions around it. However, as this video shows, having a good understanding of how this parameter works is crucial to achieve a better image quality.

  • @nookenieminen5826
    @nookenieminen5826 10 หลายเดือนก่อน +1

    Good topic Julien.
    BW is nowadays little forgotten parameter for most of the time for many radiographers. I always try to look that (too) and because I’m mainly Philips MRI-oriented, when using Siemens, always pointing mouse to BW-box to get pop up window to show WFS instead of BW. It’s much easier to remember especially when using 1,5T and 3T in different days. WFS is same, BW you have to double in 3T to get same WFS. Of course pixels are smaller in 3T so WFS is smaller then in millimeters.
    In 3T orthopedic imaging I try to use WFS about 1.6-2.0 in FS seq.- it's about 260-215Hz/Px. In T1(or seq. without FS) I lower WFS to 1-1.5 (430-290Hz/px). Or lower with BLADE/Multivane/Propeller to 0,6- 1 WFS. In the video I think the ’problem’ is in T2 FS axial-sequence where used ES (echo spacing) was 14,3ms, little too high in 3T -maybe causing blurring/motion artefacts from breathing and veins. TF 12 and ES 14,3ms totals TF-chain lenght to 171,6ms, which is ok in T2-seq, but in 3T T2 FS I would try using higher TF to 15 and lower ES about 10ms resulting TF-chain lenght 150ms, but less blurring/motion and sharper looking images.
    It’s always good if possible in PD and T2 sequences to get asked (desired) TE in the center of TF-chain (’balanced TF-chain’) -so this desired echo is pointed in the center of k-space. I use in T2 FS TE about 60-80 (or maybe TE 60 to get little more signal) and if TF is 15 the 8. echo would be 80ms and in the middle of k-space (’balanced echo train’). Echoes 1-5 and 11-15 goes in outer k-space to gain resolution (or edge detail) and echoes 6-10 (60-100ms) in the center of k-space is doing mainly the contrast. With lowered ES and WFS 2 or under without FS- less chemical shift, artefacts and blurring. In orthopedic shoulder imaging in axial sequence I prefer PD FS instead of T2FS with TF 9-13, ES 10-8, WFS 2-1.8.TF-chain must be under 130ms when TE is 30-40 to get sharp images.
    In the Juliens video wider BW400Hz/Px images in T1-coronals is in my opinion much better looking, you can see labrum and small details clearly sharper. 200Hz/Px in T1 3T is not good image with WFS about 2.15 in my opinion. In T1 cor ES was 11.ms1 and the first echo 11.1ms (asked TE 11) is pointed in the middle of K-space with high signal amplitude. Also TF 3 and ES 10 would be ok-less SAR in 3T when only 3 echoes collected in one TR. In T1 there is no point balancing Tf-chain, when only 3-4 echoes and you want TE as low you can get, so the first echo is the desired echo in the middle of k-space.
    When using AI, I think you can raise BW/lower WFS even more to levels usually used only in metal reduction sequences and get better images when AI takes noise away.

    • @everything-mri
      @everything-mri  10 หลายเดือนก่อน +1

      Wow, your detailed explanation about bandwidth and its application at different magnetic field strengths for specific body parts is truly impressive! Thank you for sharing your knowledge. I've learned a lot today and can't wait to apply your recommendations on my scanner 😍

    • @labistaytuned
      @labistaytuned 7 หลายเดือนก่อน

      It is not necessary to keep the TE at the middle of TF chain since System can automatically change the k space sampling pattern either centric or linear based on the TE selected.
      But I would avoid keeping the second echo in ETL as my effective TE as it is the combination of signal from spin echo and stimulated echo but aren’t in steady state, the point spread function could impose slight blurring in the image.

    • @nookenieminen5826
      @nookenieminen5826 7 หลายเดือนก่อน

      ​@@labistaytuned Yes, in Siemens it's automatic, in Philips you choose yourself asymmetric profile order, if you want that. But still, if you collect 13 echoes and ES is 10, then TE30 is of course possible, but a lot echoes are way over 30, so you get more T2 in image from k-space outer zones. Also later echoes has low signal, low edge detail,if varieble flips aren't use.

    • @nookenieminen5826
      @nookenieminen5826 7 หลายเดือนก่อน +1

      @@labistaytuned Stimulated echo artifact (or Fid artifact). You are right, second TE is sensitive to those. I think remember to take a notice the possibility to FID-artifact also fixes simulated echoes? So Gradient crusher or Flow comp or more than 1 nex usually fixes FID artifact and stimulated echoes at same. Nowadays in TSE imaging ES is somewhere between 6-11ms, so usually you might choose second echo at your TE in T1 or PD under TE30-seqs. I think stimulated echoes occur more if refocusing flip is low so there is longitunal magnisation left to next echoes. Very good reminder from you, thanks! Stimulated echoes can occur as artefact also in DWI imaging, and for purpose in use in STEAM-DWI, and also as artefact in TSE-imaging outside FOV as 'cusp annefact'.

    • @labistaytuned
      @labistaytuned 7 หลายเดือนก่อน +1

      @@nookenieminen5826
      True, agreed.
      If refocusing pulse aren't true 180 degree then stumulated echo is presented, probably not very problematic for t1 imaging.

  • @jasonparreno2245
    @jasonparreno2245 8 หลายเดือนก่อน +1

    Very informative
    Thank you 🙏

    • @everything-mri
      @everything-mri  8 หลายเดือนก่อน

      Thank you. Glad you found this video useful 🙏🏻

  • @fishyKRIMINAL
    @fishyKRIMINAL 10 หลายเดือนก่อน +1

    Amazing work explaining bandwidth

    • @everything-mri
      @everything-mri  10 หลายเดือนก่อน

      Thank you so much. Glad you liked the explanation 🙏🏻

  • @aneetaedwards3775
    @aneetaedwards3775 5 หลายเดือนก่อน +1

    hi decreasing the phase fourier like 7/6 will reduce the scan time and snr, how we should use

    • @everything-mri
      @everything-mri  5 หลายเดือนก่อน

      Hello! Setting the phase partial Fourier parameter really depends on what you need from the examination. For instance, if you're after high SNR images and your patient can stay still, you might want to skip using partial Fourier. However, if you’re dealing with a patient who finds it tough to keep still or you're imaging areas like the heart that are constantly moving, adjusting the partial Fourier could really help. My advice would be to go for a higher partial Fourier ratio, like 5/8 or 6/8, when you need to speed things up. This can cut down the scan time quite a bit, though it does mean a bit of a compromise on image quality and a higher chance of artifacts popping up. The 7/8 setting is a more balanced choice; it speeds up the scan a little without sacrificing too much quality, making it a great choice for those times when you need a faster scan but still care about clarity. Hope this helps!