Rare intermittent EPI artifacts: Spiking, sparking and arcing

 
Whatever you call them - spikes, sparks or arcs - the presence of unwanted electrical discharges during data acquisition can have a dramatic effect on the appearance of your EPIs and will likely result in poor or unusable data. (See Note 1.) There are many potential sources of unwanted electrical discharges - what I shall refer to as spikes for the rest of this post, regardless of the origin - in and around an MRI scanner. They can arise from within the scanner itself, or from items in the magnet room, or from items of clothing on a subject who hasn't been screened quite as thoroughly as he might have been.

Before we get to the sources, however, let's take a look at what we're talking about. Take a look at this mosaic of EPIs:

See the problem? No? Exactly! As I have mentioned several times in the past, many artifacts are best (or only) seen once the background level is brought up. Like this:

Aha! We clearly see the artifacts in this view: strange, variable patterns across entire slices.

Now, it isn't always necessary to crank the background intensity up to be able to see the effects of spikes, as we will see below. But as a general rule, the very first signs of spiking will be quite subtle and will likely be hidden away down in the noise with the N/2 ghosts and all the other crud. This is when you want to catch them, before they become intense and wreck your experiment. So, just to reinforce the point, take a look at this video and see if you can detect any anomalies in the images:

Common persistent EPI artifacts: Receive coil heterogeneity

 
The RF transmit (Tx) and receive (Rx) duties have been performed by separate coils on most commercial clinical scanners for about a decade. These days it's rare to find a combined Tx/Rx coil in-use for brain imaging, although they do exist. (We used one at Berkeley until 2008, on a Varian 4 T scanner.) The separation of Tx and Rx is generally regarded as a good thing because it means a large, body-sized coil can be used for Tx, thereby providing a relatively homogeneous transmission field over a region such as a human head, whereas a smaller (head-sized) coil can be used for Rx, thereby providing the higher intrinsic SNR that comes from using the smallest possible magnetic field detector. (As a general rule the smaller the coil, the higher its SNR close to the coil, because the sensitivity drops off with the reciprocal of distance.)

Indeed, most modern Rx coils aren't single electronic entities at all, but arrays of smaller coil elements put together in a "phased array." The entire phased array acts as a single coil only when the individual signals from individual channels are combined in post-processing. (Each coil element has its own receiver chain - preamplifier and digitizer - allowing separate treatment of signals until after acquisition is complete.) The details of these phased array coils and the combination of the separate signals aren't important at this point, although in subsequent posts they will become important. All we need to focus on right now is simply the fact that a multitude of individually received signals will be combined to produce the final MR signal. (See Note 1.) So, in this post we will consider the receiver characteristics of having multiple discrete coil elements.


Receive fields for phased array coils

Why is the modern Rx head coil a collection of separate circuits? A head-sized, single-circuit Rx coil would detect noise from the entire head, whereas redesigning the coil into a succession of small elements reduces the noise "field of view" for each element. Then, by combining the elements in an appropriate manner, the signal characteristics can be returned (as if a single circuit coil were being used) but with a reduced total noise level in the final images.

It should be relatively obvious that a small wire loop would detect signal with a localized sensitivity profile. The farther away the coil is positioned from the source of an MR signal - from a brain, say - the lower will be the voltage induced in that coil by the available magnetization. We don't need to know the particular mathematics of the receive profile - it's massively complicated for modern Rx coils in any case - so suffice it to say that there's a reciprocal relationship between the signal-to-noise ratio and the proximity of the coil from the magnetization inducing that signal. Closer is better (in SNR terms).

For brain imaging, then, it follows that signal from frontal lobe will primarily be detected by loops at the top of a head RF coil, whereas signal from occipital lobe will primarily be detected by loops at the bottom of the coil. Midbrain regions are where things get most interesting, from an electrical engineering perspective, because we need all the coil's elements combined to get appreciable sensitivity. Thus, we can state another general property of phased array coils: at the spatial scales defined by brain anatomy, a phased array coil offers a heterogeneous receive profile. How heterogeneous? is the important question.

The figure below, taken from Wiggins et al., demonstrates the SNR that can be expected from a typical brain for three different phased arrays. These sensitivity maps don't depict precisely how the Siemens product 12-channel and 32-channel head coils will perform, but we can use this comparison to give us a good idea of what we should be expecting to see in our EPIs because the general properties are consistent: the larger the phased array (i.e. the higher the number of independent elements) the smaller the individual detecting loops, the more heterogeneous the receive profile:

(Click to enlarge.)

Common persistent EPI artifacts: RF interference

 
Time to get back to the artifact recognition series of posts, all of which have the Artifacts label in the footer. RF interference (RFI), or more generally electromagnetic interference (EMI), is another one of the insidious artifacts that can be difficult to diagnose online, during an experiment, unless it becomes catastrophically bad. Your scanner is equipped with sensitive, specific tests for RFI that are used by the service engineer (and probably your physicist) to check for problems, but imaging isn't a sensitive test. Consequently, avoidance rather than diagnosis is usually the preferable option during an fMRI experiment, and a little bit of care and standard operating procedures should suffice to ensure minimal hazards to your data.

I'll begin this post with a description of the nature and sources of RF interference in the MR environment, then provide an example of RF interference in EPI time series data. Next I'll describe the sorts of things you should expect to do when you want to interface a new device, such as a button response box or a physiological monitoring unit, to your scanner as a component of your experiment. It's not - at least, it shouldn't be - a case of "plug n' play!" Finally, I'll describe a simple procedure you can follow to ensure minimal to no problems for your experiment, assuming that your facility has been set up properly.


What is RFI and where does it come from?

A nominal 3 tesla scanner is operating somewhere in the range 120-130 MHz. My scanner is parked at 123 MHz, with a magnetic field strength of 2.89 tesla. (Correct, it's only a 3 T scanner to one significant figure!) A quick glance at the FM dial on an analog radio receiver suggests immediately that the operating frequency of your MRI isn't all that different to your local broadcast radio stations. MRIs aren't the only devices operating at tens and hundreds of MHz in normal operation.

Amazing accounts of fires in and around MRIs

 
An article in the latest installment of The RADIANT is just too remarkable not to share. The article reports two MRI facility fires. In the first, the fire started away from the scanner but ended with the fire out and the magnet still on, surrounded by charred debris. The magnet couldn't be shut down (quenched) because the fire had destroyed the emergency quench circuitry! In the other incident the cause of the fire was the MRI scanner itself; arcing in the gradient cables. Read the article, look at the pictures. Thought-provoking stuff.

THIS MAGNET IS STILL ON!!!! (From http://yfrog.com/nut2nicj)

I'm hyper-sensitive to both of these scenarios, the first because we are about to move my scanner into a brand new building so I am redoing the safety training and reviewing procedures, and the second because my scanner had some serious arcing in 2010. Luckily the arcing was caught before the whole facility went up in flames. Even so... Here's the penetration panel where the gradient power lines enter the magnet room:

 

Note the charring at bottom-right, the negative terminal for the X gradient. That's the gradient used for readout for EPI so it gets by far the most use in my scanner. (FMRI is practically all we do!)

Here's the charred filter removed from the penetration panel:

And here's what ultimately happened at the gradient set, at the other end of the -Gx connection:

This picture was taken as the old gradient set was wheeled away, to be replaced with a new one. The intense heat and vibration had caused the X gradient to short out. Thankfully it was only the gradient and a filter that bought the farm. It could easily have been the entire facility!

 

GRAPPA and multi-band imaging. And motion. Again.

 

It's come to my attention that some of the latest accelerated (aka multiplexed) EPI sequences are now being made available to some sites with vendor/collaborative research agreements, a move that should catalyze their verification, testing and eventual application for neuroscience. The distribution of these pulse sequences to the wider world is great news! The potential is considerable! However, those wanting to conduct neuroscience experiments today with these zippy new tools should bear in mind the not inconsiderable risks. I want to warn you to think very carefully before taking the plunge.

Today's accelerated EPI sequences combine techniques such as multi-band (MB) acquisition with simultaneous echo refocusing (SER) and/or GRAPPA (1,2). In previous posts I've highlighted the increased motion sensitivity of parallel imaging methods such as GRAPPA. The MB family of methods also require "reference scan data" in order to reconstruct the time series images, and as such they are inherently more motion-sensitive than your plain vanilla single-shot EPI. Indeed, similar principles are used to reconstruct MB images as for GRAPPA, and the basic motion sensitivities are the same, i.e. motion during the reference data acquisitions will contaminate all images in a subsequent time series, while motion after the reference data but during the (accelerated) time series will lead to mismatches and spatial artifacts that will degrade temporal stability. In short, using these accelerated sequences is akin to sharpening the motion sensitivity profile of your experiment, and you will need to ensure a high degree of subject compliance to get good data.


Plan, then scan.

Now, I'm not suggesting you dismiss out of hand these sequences for your research. I am suggesting that you apply a lot of forethought, taking the time to consider several important factors. I've written before about evaluating pulse sequences that are new (or new to you). Your first task is to determine whether you even need a fancy, partly validated, highly risky pulse sequence to answer your neuroscience question. If the answer isn't a resounding "yes," why take the risk? Next, you should ask yourself how the pulse sequence should be set up to provide the optimum data. For instance, do you know which slice direction is best for minimizing motion sensitivity and/or receive field bias (g-factor) for the multi-band sequence? And do you know which RF coil to use, and why? If you can't establish your experimental setup based on sound principles that's a suggestion you either don't have the expertise yourself or you aren't collaborating with someone with the requisite expertise. (Me? I could guess, but that's about it! Without doing a validation study of my own I'd be winging it. Which is kinda my point!)

Please don't just go download and use the latest and greatest technique because it's new and cool. I've seen this movie before, and ninety nine times out of a hundred it ends in tears. Please put some justification and logic into your choices before you go and spend hundreds of hours and thousands of dollars finding yet another way that motion can confound an fMRI experiment. Eyes wide open!

__________________

References:

1.  S Moeller, et al. "Multiband multislice GE-EPI at 7 tesla, with 16-fold acceleration using partial parallel imaging with application to high spatial and temporal whole-brain fMRI." Magn. Reson. Med. 63, 1144-53 (2009).

2.  DA Feinberg, et al. "Multiplexed echo planar imaging for sub-second whole brain fMRI and fast diffusion imaging." PLoS ONE 5(12), e15710 (2010).

GRAPPA: another warning about motion sensitivity

 
I wrote a post in May last year to highlight the enhanced motion sensitivity of GRAPPA-EPI compared to single-shot EPI for fMRI. Paul Mullins and I had also discussed the use of GRAPPA for resting-state fMRI in the Comments of an earlier post. The literature is still fairly quiet on the adverse effects of GRAPPA for fMRI although, as I noted in the May post, there are one or two reports of reduced fMRI sensitivity when using parallel imaging, some of which might be attributable to motion (whether it was diagnosed as motion or not in the published work).

In the May, 2011 post I explained the two types of motion sensitivity that plague GRAPPA in its usual incarnation for EPI time series acquisitions. The first type - motion contamination of the auto-calibration scans (ACS) - might be mitigated by vigilance and a suitably resilient task script, e.g. one that uses plenty of null events at the start of the acquisition, before the first real stimulus is presented, to give the operator sufficient time to evaluate the images being generated with the current ACS and decide whether or not to stop and start over. This approach is no guarantee that motion won't have contaminated the ACS, but simple tactics like this can help avoid the worst effects of motion during the start of the run.

The second type of motion is that which happens after the ACS and during the (under-sampled) time series itself. This problem is one of mismatch. Displacement of the head from its position during the ACS acquisition can lead to spatial errors in the current image volume. Thus, whilst attaining motion-free ACS might be considered essential for fMRI, maintaining proper matching of the ACS to the under-sampled time series is also important. The bigger the mismatch the more likely there will be a penalty in statistical power for the time series.

In this post I want to tackle the issue of non-head motion in the scanner, and its effects on GRAPPA-EPI images. This investigation was motivated by one of my users who reported seeing occasional "banding" in a study that had used GRAPPA-EPI. The traditional evaluation of head motion suggested that the subjects weren't moving very much, so I started looking into other possible instabilities. I was quite surprised just how sensitive GRAPPA-EPI can be to small perturbations, as you will shortly see.


A quick review of some brain data

Let's begin by looking at one of the problem GRAPPA-EPI data sets from a human subject. The acquisition specifics are as follows: 12-channel head RF coil on a Siemens Trio/TIM scanner, GRAPPA with R = 2, reconstructed matrix = 96x96, FOV = 224x224 mm, slice thickness = 3 mm, 10% gap, interleaved sagittal slices, flip angle = 90 deg, TR/TE=2000/26 ms, echo spacing = 0.8 ms, readout bandwidth = 1408 Hz/pixel.

Here is a cine-loop through the raw data:

New user training guide/FAQ

I've just uploaded a new user training guide/FAQ that we use at Berkeley to initiate newbies into the ways of the dark side. It is Siemens-specific, for a Trio/TIM.

As last time, the guide is a bit rough. Sorry for English-isms and typos. It's worth exactly what you pay for it. It's free. Use and abuse it however you like. It's a Word document so that you can reorder things, add your own notes, etc. I would appreciate constructive feedback, especially if you find mistakes or have suggestions to improve it, but there's no need to ask permission to use it, change it, replicate it, sell it...

The most recent version of the training guide/FAQ is available from this web page:

http://bic.berkeley.edu/scanning

Locate the file attachment towards the bottom of the page, it's called 3T_user_training_FAQ_08Mar2012.doc. The most recent contents and a list of changes since the last version (April, 2011) appear below.


Caveat emptor.

The document is only a component of user training, don't expect to learn how to scan by reading it! Rather, use the tips to extend your understanding, refine your experimental technique and so on. Note also that this document is for a Siemens TIM/Trio (with 32 receive channels) and running software VB17. There may be subtle or not-so-subtle differences for the Verio and Skyra platforms, for software VB15, VD11, etc. so keep your wits about you if you're not on a Trio with VB17!

You may have local differences, e.g. custom pulse sequences, that allow you to do things that contradict what you find in this user guide. Talk to your physicist and your local user group before taking anything you find in this guide/FAQ too literally.

Finally, you wont find many (any?) references in this guide/FAQ. It's for the training of newbies, not a comprehensive literature review! If you are seeking further information on something I mention in the guide and you can't find a suitable reference yourself, shoot me an email and I'll do my best to point you in a useful direction.

 --------------------------------------

Update Notes (8^th March, 2012):

• Updated with new operating modes available under software syngo MR version B17.

• General tweaks to improve readability.

• Further recommendations on using the 32-channel coil for fMRI.

• Added a description of the new AutoAlign procedure, AAHScout.

• Added a new section: “I have an existing protocol that uses the old AutoAlign (AAScout). How do I get and use the new AutoAlign (AAHScout)?”

• Added a new section: “I want to add a new acquisition and acquire exactly the same slices as this other EPI acquisition I just acquired. How do I tell the scanner to do that?”

• Extended the discussion on the relative merits of PACE versus using an offline realignment alone, in the section on the ep2d_pace sequence.

• Fixed a typo concerning the slice ordering for descending slices.

• Added a new section: “What is a field map and how does it fix EPI distortion?”

• Added a new section: “I want to try to fix my distortion with a field map. What do I need to acquire?”

• Updated the sections on partial Fourier for EPI, noting that Siemens simply zero fills the omitted portion of k-space rather than doing a conjugate synthesis.

• Extended checklists.