Introducing MRI: Fast Gradient Echo and Echoplanar Imaging (38 of 56)

www.einstein.yu.edu - The thirty-eighth chapter of Dr. Michael Lipton's MRI course covers Fast Gradient Echo and Echoplanar Imaging. Dr. Lipton is associate professor radiology at Albert Einstein College of Medicine and associate director of its Gruss Magnetic Resonance Research Center.

[@effiongjay373]

I'm so glad the last part about t2* was clarified. And I wish I can enroll in an mri class from Nigeria. There's so much wealth of knowledge I just want to tap into.

[@jalapeno1171]

First, fantastic series and great explanations. Two questions: 1. For the fast gradient echo sequence, is signal acquired during the first gradient echo? Or is there a first 'dummy' gradient that dephases the spins so that an increasing signal is acquired during the next gradient? Technically, as it is explained here, the first acquired signal is decreasing during acquisition. 2. Am I understanding this correctly that in this EPI sequence, an entire 256 by 256 image is acquired in 1 - 2 ms? So that means that in every 2/256 ms, 256 data points are collected? That is quite incredible.

[@sayimgokyar362]

Hi, as far as I understand from the lecture, first 'dummy' gradient lobe causes dephasing and then the following read-out gradient creates refocusing. Let's say in positive direction. The nice part is, you do not need to repeat this 'dummy' gradient again for the second phase encoding line since the previous readout gradient lobe will be considered as the dephasing lobe of the following readout gradient. 1. Yes the signal is decreasing with T2*. That's why it is recommended to fill up the k-space from center to periphery. 2. This very usual at the moment :). I think the seminal paper reported EPI for the first time in 1991 is this: www.mriquestions.com/uploads/3/4/5/7/34572113/stehling_epi_science_1991.pdf Acquisition times are given by a couple of hundred milliseconds in 1991!

[@surajsonwane517]

very nice explanation for clearing all the concepts about the echoplanar imaging......but during the spinecho echoplanar imaging explanation in the later of this lecture there was no mention about the 180 degree pulse which we shoud give after every acquision of the echo that is neccessary, because without the 180 flip the image will be only the T2* weighted and there will be no difference between the T2 and gre image???? that is the 180 degree pulse should be given before every time we image.

[@lalithsharan4734]

In gradient echo imaging, we measure T2* , so the 180 RF is not required

[@shashwatpriyadarshi972]

Are all gradients like slice selection,phase encoding and frequency encoding applied simultaneously or one after the other? How is residual transverse magnetization refocused in unspoiled gre imaging? Why is inhomogeneity due to local magnetic field not refocused by gradient?

[@shashwatpriyadarshi972]

Bro,what is the rationale behind blips being applied before each frequency encoding gradient? Why are there two lobes during frequency encoding gradient application?

[@shashwatpriyadarshi972]

When Dr.Lipton starts to explain the girl epi ,then during the explanation he draws a frequency encoding gradient with opposite polarity in the beginning of pulse sequence.can you explain why?

[@sayimgokyar362]

@@shashwatpriyadarshi972 @Shashwat Priyadarshi Hi, - I think you should apply slice selection gradient during RF excitation. Otherwise, all of the volume will be excited. Phase and frequency encoding gradients can be applied simultaneously (for example spiral k-space trajectories) as long as you have an appropriate reconstruction methodology. - Residual transverse magnetization is not refocused (and I see no reason to focus it since the phase encoding will shift k-space data points to another place). - local magnetic fields (due to RF field) and gradient fields are two completely different waveforms and one has nothing to do with the other. A gradient waveform is used to create a frequency or phase offset among consecutive spins, where as the RF waveform is used to rotate spins with a predefined frequency (e.g., 64 MHz for 1.5T). I hope these answers would help!

[@flameeverlasting]

"There is no TR" 🤯