OPM-MEGs - New, Precise Brain Scans For Epilepsy - Christine Embury, Young Epilepsy, UK
Post-doc researcher Christine Embury shares with us her research on the cool OPM-MEG technology for brain imaging - its advantages (particularly for children with refractory epilepsy and considering brain surgery!) and how it could improve patient outcomes and our understanding of developmental trajectories in paediatric epilepsies! Lastly, Christine and Torie chat about the cost-effectiveness of the device, its benefits from an energy usage/climate change perspective, and how both patients and clinicians can get involved in the research!
Reported by Torie Robinson | Edited and produced by Carrot Cruncher Media.
Podcast
-
00:00 Christine Embury
“You can also make custom arrays that go down the spine or look specifically at different regions (if you know maybe where something might be coming from, you can make a denser array there). So, they're really flexible in that way. They can get closer to like a kid's head, which is really nice.”
00:14 Torie Robinson
Fellow homo sapiens! My name is Torie Robinson, and welcome to, or welcome back to: Epilepsy Sparks Insights.
Now, loads and loads of babies and children have refractory/uncontrolled focal epilepsies, and it can take AGES for for them to get the scans and info needed for their neurologist to understand it properly (and therefore treat it in the most effective mannar - which is, actually, often neurosurgery.
Well, right now, we have a cool team who are looking into OPM-MEGs (which is a type of brain scan that they will explain properly!) which are being combined with EEGs(!) to basically, give us 2 in 1 and save heaps of time in preparing for neurosurgery!
01:02 Christine Embury
I'm Christine Embury. I'm a postdoctoral researcher down at Young Epilepsy. I've been working in MEG in general since about 2016 doing my PhD on the effects of diabetes on the brain. So, always been really interested in like the clinical applications of the technique and how we can really improve people's lives like actually using it.
Yes, I've been down at Young Epilepsy for a little over a year. We're kind of running a study looking specifically at children with epilepsy using OPM-MEG, which is kind of a new technique. So, yeah, hopefully I get to tell you more about it today.
01:38 Torie Robinson
It sounds like opium MEG - opium the drug (but it's not!). What does OPM stand for?
01:45 Christine Embury
So, it's called an Optically Pumped Magnetometer. And then MEG is Magnetoencephalography. So, an optically pumped magnetometer is basically, just like a kind of sensor; a way to actually sense the magnetic waves from the brain. Magnetoencephalography is literally the writing of magnetic waves from the brain. If you look back at like, I don't remember if it's Greek or whatever, but if you actually go and break down the word that it at least it just means that we're looking at the magnetic waves specifically from the brain. And so, basically, this is a new technique using magnetoencephalography in a way that we can apply directly to the head. So, traditional cryogenic MEGs require the sensors to be bathed in liquid helium to keep them at, like, super-conducting temperature. And all that means is really that they have to be really cold. So, they have to be kept in a dewar inside a big helmet and an array that's static (it only stays in one configuration). And now we have sensors, basically, that can be done at room temperature; the lasers themselves pump them to a little bit above what body heat basically is. They get a little bit warm - but they're…they’ve got like nice little coverings and things, but they're inside a I don't know, little black cube about the size of a Lego brick. There's 2 lasers in our sensors, but they can go up to 3. So, you can get lots of different directions of the signal from one little sensor, and then you can put them in any kind of array that you need. So, you can put them in… we have different sized helmets for different head sizes, but you can also make custom arrays that go down the spine or look specifically at different regions (if you know, maybe, where something might be coming from; you can make a denser array there). So, they're really flexible in that way. They can get closer to, like, a kid's head, which is really nice.
03:33 Torie Robinson
And different shaped heads as well!
03.34 Christine Embury
Exactly!
03:34 Torie Robinson
Right?! I imagine because some people have like…. I was reading this article, short article about how it's outrageous that some people have flatter heads and it's really difficult to get, for instance, a cycling helmet to fit your head. Right.
03.47 Christine Embury
Exactly!
03:47 Torie Robinson
So, I guess this is one of the benefits [of] what you're talking about, right?
03:51 Christine Embury
Exactly. And while we have these helmets that are at least a little bit more static - like a bicycle helmet - at the moment, the whole idea is that these things are ever evolving and they've got new ones all the time that can, you know, basically, transform to the shape of your head; closer to, like, what an EEG normally would, but, still be able to track where each one of the sensors are, relative to the head. So that you can go back and actually get the spatial resolution that you need with MEG. So, they're really cool and really new. They've only been around since, I want to say around 2016, 2015, something like that. And so, it's just really fun to kind of see how they're already being taken so much more closer to the clinic than, really, it took cryo-MEG to do in decades. So, it's very cool.
04:37 Torie Robinson
Yeah, and they are evolving so, so quickly right if you see the timeline, like; you say from 2016 to now, it is nuts how far we have come.
04:47 Christine Embury
Yeah!
04:47 Torie Robinson
Just put simply: how does this device compare to… and say, MEG on its own, and MRI, and EEG on its own… and how are you combining the former two?
05:01 Christine Embury
Basically, we're looking at the functional activity of the brain; so, how the brain is behaving while we show a different stimuli - or even at rest. And so, we don't really get a structural image (something like MRI), you usually have to get an MRI and overlay that data on it (something called “co-registration”), or you can use a template data, that kind of thing. So, really you're looking at just the functional data, just kind of, like, EEG. So, EEG is the same kind of thing. The kind of advancement here is that you actually, like; EEG looks at the electrical signals of the brain; so, electrical activity in general gets spread across all kinds of surfaces that are conductive. So, things like your blood vessels, your skull and your scalp, all those things can distort that signal and kind of affect how well we're able to tell, spatially, where that's coming from. And so, magnetoencephalography on the other hand: it isn't really impeded by any of the skin, skull, scalp, or anything. It's a 1-to-1 kind-of direct measure of the neurophysiology underneath of it. So, the neural activity that's happening: you're actually still measuring at the sensor in the same way. It's not like the bold signal and functional MRI (where it's the blood that's actually flowing to that region and that inherently is part of the activity); it's not like that. It's actually all the presynaptic changes in the electrical potentials on the neurons. So it's really cool that it's a direct measure. So, you're able to get down to something like a millimetres level precision of the thing when you're trying to localise the signal. It's very cool that way.
06:33 Torie Robinson
And, also, I believe MEG, if you compare it to EEG for instance, it can access tissue that no way an EEG can.
06:41 Christine Embury
Yeah, yeah! So, as long as it's well aligned to the array - some of…
06:46 Torie Robinson
Right.
06:46 Christine Embury
…the signal generally falls off of the square of the distance. So, the further away it is from the sensors, the harder it is to see, but we have lots of techniques to kind of cancel out noise around it, to get down to those deeper sources that you couldn't necessarily get with other techniques. So, there's lots of ways that we can do that; whether that's upping the signal to noise by giving them a specific task that says “Ok, now activate that hippocampus over and over and over again” So, we can find that signal, or, it's just recording for longer, making sure you have a really nice array that can, you know, capture all the signal possible from that area.
07:19 Torie Robinson
Given that these are fancy machines and the combination of machines and not the cheapest in the world, right? How accessible are they? And how accessible is this type of scanning?
07:34 Christine Embury
Not only is it about the cost, but also just generally where they are in the world. So, MEG has traditionally been used, clinically, for, like, epilepsy surgery planning, mapping out the eloquent cortices around just the epileptic foci, but also the epileptic foci. So, generally, it has some clinical applicability, but there aren't as many centres doing that - there's some back in the States, certainly there's some big centres doing that, [but] In the UK, there's only one clinical centre, actually, that is available - up in Aston or Birmingham Children's (they kind of share that facility). And so, really, it's about “How do we get these out there to more places?” and the machines that we have here today are actually much less expensive than traditional cryogenic machines. Not only are the actual sensors themselves modular (and so, you can buy different sizes of arrays and things like that to make it cheaper), but also, the rooms they've now made lighter and [they] have different kinds of shielding in there (so, like, more active shielding rather than needing all the passive, really heavy, new metals to build them. So, you can have lighter rooms and that takes away a lot of the cost - and somewhat of the [carbon] footprint as well. ‘Cause it usually had to have them on a ground floor or basement of a place, ‘cause they're really, really heavy, and now you can kind of move them around; they're a little bit more flexible about where you can actually put them). And so, really, the whole system itself costs probably about half of a traditional cryogenic MEG. And that's not…
09:05 Torie Robinson
Which is about how much?
09:06 Christine Embury
Uh, I think the system costs probably about a mill and a half?
09:11 Torie Robinson
And that's the cheaper one that we have now compared to the older one?
09:14 Christine Embury
Exactly. And so, all in all, you can get, I mean, you can do that in different ways, like you can make smaller arrays, you can get more sensors, but lighter room, like, there are lots and lots of configurations and they're even trying to make even lighter rooms as they go forward and smaller rooms. So, like, our room’s a mid-sized room (for what they offer for one of the companies), but there are lots of different arrays that can be even cheaper than this, which is really cool. And not to mention that it also costs less in general for operating costs. So, like, traditional cryogenic MEGs require those cryogenics to be replenished; some of them require using a helium recycler (that constantly keeps that recycled); which depends on whether or not it's an aftermarket or actually put in at the same time as the MEG and how efficient it can be. So, sometimes you have to still fill them up every week or 2weeks. Sometimes you don't have to do another fill until a whole other year, but it uses things like liquid helium, which is a finite resource. And so, it's really hard to get a hold of it and it's really expensive and a bit dangerous actually to do. I’ve done helium fills myself; they can be quite dangerous.
10:24 Torie Robinson
Gosh!
10:24 Christine Embury
So, anyway, it doesn't require any of that! I can actually turn the whole system off and that's not a thing you can ever do with the cryogenic MEG. You can warm it up and then get rid of all the helium and then refill it with a whole bunch of helium, but this one, yeah; I can just go turn off each one of the switches and it's off for the day, and so,
10:43 Torie Robinson
Ahh, that's cool.
10:43 Christine Embury
…and then, resource wise: it uses quite a lot less and it causes like it's a lot less operator cost in general. You have to still pay for all the expertise to run it and some, like, annual contracts to keep it up, but really that's your cost. It's not, like, finite resources that are obviously, constantly needing replenishing, and so, in that way it's quite a bit less expensive and hopefully more accessible to other facilities to be able to use.
11:08 Torie Robinson
So, we have a good business argument here too,
11:11 Christine Embury
Yes!
11:11 Torie Robinson
…as well - as for why they should be… and a climate change argument…
11: 14 Christine Embury
Exactly!
11:14 Torie Robinson
…perhaps one may say more importantly: climate change. Because I have heard, you know, clinicians (especially neurophysiologists) talking about the, you know, the… almost a bit of a guilt feeling of how much the MRIs do contribute to climate change.
11:31 Christine Embury
Exactly.
11:32 Torie Robinson
I mean, it's a tiny bit compared to everything else(!), but, this can… but significantly for a medical device it's crazy.
11:40 Christine Embury
When you have to look at the whole medical ecosystem in general, it's quite wasteful, so, anywhere that you can kind of pull back some of that, the better.
11:47 Torie Robinson
So, there's a good argument for any hospital leaders out there, universities, you know!
How could it potentially benefit patients? Like, I'm thinking, could it literally save them money, provide earlier, more effective care, things like that?
12:04 Christine Embury
Yeah, so, I think one of the best use-cases, basically, for it - and why we're doing our current study - is, basically, we can get into younger kids. So, generally, with cryo MEGs, you either have to have a really specialised system for children (and even that can be only down to certain ages and those are really rare and you have to get a whole special equipment) where[as with] these I could get down to babies and put a different array (like a smaller array) on them and actually still be able to use it. And so, really, you could save your time and work up because you can get faster to that surgery pathway and people who need to kind of get there because that's a real, you know, actual curative thing…
12:46 Torie Robinson
Yeah. For many at least.
12:46 Christine Embury
…rather than having to… yeah, exactly… instead of having to just try different medications and things. It's one of those things where we can get down to younger kids and maybe actually scan them younger and be able to get them on that clinical surgical pathway earlier and easier.
13:04 Torie Robinson
Totally, and I'm just thinking, sorry, as somebody who grew up as a child (potentially still am), as a person with an epilepsy, and having noticeable cognitive, almost, I don't know if you would even call it regression, but limitations over the years as a result of refractory epilepsy. I can see for a child and a baby; if you could minimise that by treating them as early as possible (because you know how to as a result of this scanning) and you could prevent the regression and then the onset of other morbidities which often accompany the morbidity of the seizures, right?
13:36 Christine Embury
Yeah, yeah, and that's definitely, like…; part of our overall research aims is to kind of track how these play into developmental trajectories. And we're doing not just the work to try and find the epileptic foci, but also doing some of that pre-surgical mapping and looking at things like language function and things like that to really see what that's like in each of these kids' brains. And so, I think it is important not just to do… just what is exactly clinically applicable, but also to do that research to look at developmental trajectories and things like that; to really try and make an impact on the overall outcomes for them. Because it's not just about just containing the epilepsy, it's also about so, much other different cognitive aspects, like you say.
12:17 Torie Robinson
And I imagine people, you know, like, with… it could be a distinct benefit to people with DEE like developmental… and I've forgotten the word now but DEE, hahaha! It's because I'm feeling the pressure! Developmental and Epileptic Encephalopathy! There we go! Hahaha!
14:30 Christine Embury
Exactly! You got it!
14:31 Torie Robinson
There I got it, I got it, finally!
How can people help in your research if they want to? Clinicians, patients, parents?
14:41 Christine Embury
They have to get in contact with their clinicians - so, we have to have all of our stuff through NHS trusts and other clinicians and stuff, so we can exchange clinical data back and forth. So, we need things like previous MEG or EEG scans, things like that, just to make sure our technique actually matches up.
14:56 Torie Robinson
Ok.
14:57 Christine Embury
So, really, not direct contact with the patients, but more, so, like, through other clinicians and providers. So, if there's any clinicians and providers who are interested; we're always looking for new sites to add onto our ethics to kind of contribute to this and be, like, a patient like identification centre, that kind of thing. But yeah, otherwise just getting your clinicians involved in research, like, I know that this is still, you know, at early stages; we're still doing things to develop it, to really make it a tool that maybe can be used for clinical things in future. It takes investment on everyone's side, so, really getting clinicians involved in this kind of step is helpful.
15:35 Torie Robinson
And just to clarify for families that want to be involved, this isn't something that will benefit their children directly, right? This is for research alone at the moment - for people of the future, right?
15:47 Christine Embury
Exactly. So, really, this isn't like a clinical tool yet. The idea is that we have to, you know, do this work so that we can make sure that it is safe and effective the same way that we say that it is, that we think that everything works. So, we have to do, you know, some of that work to compare it back to EEG and things like that to really get it on that pathway first.
16:08 Torie Robinson
Thank you so much to Christine for giving us a glimpse into her research - and making it fun and understandable!
Do check out Christine’s work on epilepsysparks.com (on the research page) or torierobinson.com (where you can also access the podcast, the video, and transcription of this episode).
If if you haven’t already, please don’t forget to like, comment, and subscribe to our channel, share this episode with your friends/colleagues/family members(!) and see you next week!
-
Christine Embury is a postdoctoral research fellow at Young Epilepsy utilising OPM-MEG in clinical contexts. She earned her PhD in Psychology and Neuroscience in 2021 from the University of Nebraska Omaha examining the impact of diabetes on the brain and cognition using MEG. She harnessed the application of the technique across the lifespan with and without additional disorders examiningneurophysiological responses and their relation to behaviour and cognition. She has over 40 co-authored publications in high-impact and field-renowned journals, including Developmental Cognitive Neuroscience, Neuroimage, Aging, Cerebral Cortex, and Diabetes, with over 500 citations.
Her current work focuses on imaging childhood epilepsy using OPM-MEG to determine the method’s capability of measuring epileptic responses and mapping eloquent cortices. The goal is to assist in building a clinically viable set up to provide a more accessible neurophysiological diagnostic and treatment-guiding measurement to children with epilepsy.
-
LinkedIn: christine-embury
Young Epilepsy Determining the utility of OPM-MEG in a clinical context
Young Epilepsy: power-research-joint-research-reception
ResearchGate: Christine-Embury