0:02

Hello and welcome to Noon Conference, hosted by modality

0:05

Noon Conference connects the global radiology community

0:08

through free live educational webinars that are accessible

0:11

for all and is an opportunity

0:12

to learn alongside top radiologists from around the world.

0:16

You can access the recording of today's conference

0:18

and previous noon conferences by creating a free account.

0:21

Today we are honored to welcome Dr.

0:23

Roland Lee for a lecture entitled Imaging of Epilepsy.

0:27

Dr. Lee completed his radiology residency at Brigham

0:30

and Women's Hospital and Harvard Medical School,

0:33

an MRI fellowship

0:34

with Dr. William Bradley at Long Beach Memorial

0:36

and Neuroradiology Fellowship at UCSF.

0:40

He's currently chief of Neuroradiology

0:42

and director of MRI at VA San Diego,

0:45

and director of MEG at UCSD VA San Diego.

0:49

At the end of the lecture, please join Dr. Lee in a q

0:52

and a session where he will address questions you

0:54

may have on today's topic.

0:55

Please remember to use the q

0:57

and a feature to submit your questions so we can get to

0:59

as many as we can before our time is up.

1:02

With that, we are ready to begin today's lecture.

1:05

Dr. Lee, please take it from here.

1:08

Good. Um, morning and good afternoon.

1:10

Uh, to the audience, thank you for attending, um,

1:13

this lecture, I'm gonna talk today about, uh,

1:15

magnetoencephalography, um, basic concepts

1:18

and applications to epilepsy.

1:21

Um, I know that, uh, many are not that familiar with,

1:25

with MEG Mag Ence, so I'm gonna go, uh, tell you about it.

1:28

I think it's a very, um, useful modality, um,

1:32

clinically useful, um, especially for epilepsy.

1:37

Um, also for, uh, presurgical mapping.

1:41

Um, there are other, uh, utilities of MEG such as diagnosis

1:45

of concussion because you're looking at brain function,

1:47

not structure like MRIs.

1:48

So it's actually good for many things,

1:50

but it's, it's tailor made for epilepsy.

1:56

Okay, so this is what the, um,

1:58

a typical MEG machine looks like.

2:00

Um, it's a helmet, uh, that, uh,

2:05

the patient, uh, sits in the chair,

2:06

head fits tightly in this or else the gantry can tilt.

2:10

Patient can lie down, and the head will again, fit in this.

2:12

There's a bunch of sensors in here in this particular

2:15

1, 306, um, channels,

2:17

which measure the magnetic field of the brain.

2:24

Okay? So, MEG, um, is completely noninvasive.

2:28

Uh, unlike MRI, it doesn't put magnetic fields

2:31

or, um, radio frequency fields into you.

2:33

It only measures your, um, magnetic fields of your brain,

2:37

sort of like EEG

2:38

or sort of like measuring your temperature

2:40

with a thermometer, um, in your, um, in your mouth.

2:43

Uh, it localizes the electrogenic, uh, trigger zones, um,

2:48

which really helps the, uh,

2:49

surgeon pre surgically if they're going to, uh,

2:52

take out, uh, trigger zone.

2:54

It also can, uh, localize eloquent cortex

2:57

so you don't damage the patient when you do surgery.

3:00

Um, it reduces dependence on invasive studies

3:02

before MEG to, uh, get, um, a a detailed look at the brain,

3:08

uh, function.

3:09

You would have to take off part of the skull

3:12

and put, um, electrodes on, for example.

3:14

Um, uh, it also optimize the accuracy invasive study.

3:18

So nowadays there are, um, many centers are doing, uh, uh,

3:23

um, stereo, uh, they're doing, um, uh, EEG with,

3:28

uh, very small electrodes.

3:30

And so, um, this way we can, uh, guide them where

3:35

to put those electrodes.

3:37

Uh, therefore, it reduces the time in the operating room,

3:39

therefore, reduces cost of epilepsy surgery workup,

3:42

and is shown in literature improve outcomes

3:45

of epilepsy surgery.

3:46

Also, it is clinically accepted.

3:48

It's not research insurance does pay for MEG workup.

3:52

Um, here's, for example, the American Academy of Neurology.

3:55

They have this, uh, policy, which does, it's indicated

3:59

for presurgical evaluation of epilepsy, um, also for, uh,

4:02

surgery, um, preoperative, uh, evaluation.

4:06

Um, yes, MEG is not the first, uh, order of test.

4:11

It is not a standalone test, um,

4:13

but it is useful, um, in problem solving in patients, um,

4:17

where the MRI alone cannot give the diagnosis

4:20

of where the lesion is.

4:22

Okay. Okay. So just telling you about how MEG works.

4:27

As you know, the way the brain works is

4:29

that there's electrical currents, um, jumping from, uh,

4:33

one neuron to the other.

4:35

And, um, the neurons are arranged pyramidally, um,

4:39

in about one cubic millimeter of tissue.

4:41

That's about a hundred thousand, uh, a million neurons,

4:45

and they're all arranged together.

4:46

So they're a current sum.

4:47

So we can see the sum of that many, uh, parametal cells.

4:52

Um, you can see the electric fields by putting electrodes,

4:56

you know, on the skull,

4:58

or if you take off the skull in the brain,

5:00

but you can see the, uh,

5:01

magnetic fields without removing the skull.

5:04

And that's because the, um,

5:06

magnetic fields are not disturbed at all by skull,

5:09

whereas electric fields are very dis, um, you know,

5:11

distorted by the skull.

5:12

These detectors, um,

5:14

to detect the magnetic fields are called super connecting

5:17

quantum interference devices invented by, uh, uh,

5:20

professor Josephson who won the Nobel

5:21

Prize for inventing them.

5:23

Um, when we do this, we can get a measurement

5:27

of the magnetic field at all points

5:29

around the head like this mathematically.

5:32

Then you can calculate

5:33

where in the brain at any given millisecond,

5:36

the ACT activation is in the brain.

5:39

So again, the patient can be lying down or sitting up,

5:42

and you can measure the patient, uh, sleeping or awake, um,

5:47

because the magnetic fields are weak.

5:48

You need a very strong, uh, thick shield shielding room, um,

5:52

to screen out the, uh, extraneous magnetic fields.

5:55

For example, right now, you're getting a lot

5:57

of electromagnetic radiation from cell phone towers, tv,

6:00

radio, and so forth.

6:02

Those are all screened out, and must be

6:04

because the magnetic fields of the brain are so weak.

6:06

So, again, here's the sensor picking up the signal.

6:09

Um, here's all the tissue that the, um, magnetic

6:13

or electric field has to travel through.

6:15

And again, when I put the electrode on the skull, uh,

6:18

on the scalp, all these tissue distort the electric field.

6:21

None of them distort the magnetic field, which, as you know,

6:23

is only, um, distorted by metal.

6:26

Again, this shows the how, uh, weak the brain activity is.

6:31

Uh, brain noise algorithms, uh,

6:34

which is one of the strongest signals.

6:36

Um, standard noise that we live in is this much,

6:39

but if I have epilepsy, the noise, the signal's very high.

6:44

So, um, clinically reimbursed applications, again,

6:46

localizing interal spikes for epilepsy

6:49

and are also, um, localizing major functional,

6:52

eloquent cortex, uh, pre-surgery, for example,

6:56

at our center, um, it's an outpatient center.

6:59

Um, because we don't give any medications,

7:02

we don't need any sedation, um,

7:04

it's ver it's completely silent.

7:06

There is no noise unlike an MRI scanner,

7:08

and there's no claustrophobia

7:09

because there's like putting on a fist football helmet

7:11

or a bi bicycle helmet, so you never get claustrophobia.

7:15

Um, there's never any noise to distract the patient.

7:17

They're just sitting in a chair,

7:18

like in their classroom or sleeping.

7:20

Um, if you want to, uh, increase the yield of spikes, um,

7:23

we can do simultaneous EEG, um,

7:26

because there's no applied fields at all.

7:28

It's as safe as taking a temperature, um, you know,

7:32

with a oral thermometer.

7:35

Um, if they have a, uh, vagal nerve stimulators,

7:38

we can ma um, do MEG.

7:40

We just gotta turn it off, uh, deep brain stimulators.

7:43

There may be too much artifact,

7:44

but again, if there's any questions, um,

7:46

the patient just comes in and we just, uh, do a, a noise run

7:49

to look for the, if there's too much noise or not,

7:52

but also even with some noise mathematically,

7:54

you can get rid of a lot of the noise

7:56

and just subtract it out.

7:57

There actually are CPT codes in the United States that, uh,

8:01

um, uh, do pay for these, uh, um, MEG modalities.

8:07

So just how it works.

8:08

Let's say I stimulate your, uh,

8:09

median nerve at time equals zero, um, with a, uh,

8:14

a little shock to your wrist.

8:15

Then you can see millisecond by millisecond the response.

8:18

So 20 milliseconds

8:19

after I, um, shock you,

8:21

I get a magnetic 20 millisecond response.

8:24

And here's a, a map of all the sensors

8:26

around your head, and I can see it.

8:28

This is right, this is left.

8:29

If I stimulate your right median nerve,

8:31

you're gonna get left brain activity.

8:33

So, if I say, let's assume it's one point in time that, uh,

8:38

is your, uh, point of stimulation, uh, which is valid

8:42

for media nerve stimulation

8:43

or for epilepsy, you can very easily

8:46

calculate where that point is.

8:48

So here I see that the, um,

8:50

median nerve stimulation is on the posterior

8:52

bank of the central sulcus.

8:54

I can, uh, move my index finger and I look 20 milliseconds

8:57

before my finger moves timelock to that.

8:59

I can see it's exactly on the anterior bank

9:01

of the central succus, as we all know, similar,

9:03

I can stimulate the, uh, tibial nerve

9:06

and so forth, so I can, uh, localize the, um, somatosensory

9:11

and motor cortex in, uh, normal patients

9:13

and patients with tumors.

9:15

So this can help the surgeon if

9:16

he wants to remove this tumor.

9:17

He knows he wants to approach it from the front instead

9:20

of from the back to avoid hurting the patient.

9:23

So this is, um, been confirmed numerous times.

9:28

Um, in everybody's lab, you do it on your, your own,

9:30

and there's literature on it.

9:32

So you can localize lesions such as, uh, the, uh, cavernoma

9:37

localize where the, um, motor strip is.

9:40

You can check it with functional MRI

9:41

and also intraoperative electrocorticography.

9:44

So we know how it works. We also have, um, multiple, uh,

9:49

institutions, um, can, uh, use paradigms

9:53

to localize brokerage area

9:55

and, um, our, our lab, for example,

9:58

and, uh, other labs across the country.

10:00

Here's, uh, sick kids in, in Toronto.

10:03

So we know that we can, uh, localize, uh, brokerage area

10:07

and Wernicke's area in MEG reliably.

10:10

So, epilepsy, the current situation here, um,

10:14

this is from one of my, uh, uh, friends who's a, uh,

10:17

a neurologist specialized in epilepsy.

10:19

About 25 to 38%

10:22

of patients epilepsy will not become seizure free with,

10:24

with, uh, medication alone.

10:27

The good part is about 70%

10:30

of people are treated well with medications.

10:32

So those patients we never see need to see those.

10:35

They just take their pills, they're fine, and that's good.

10:38

But for that, about 30% of patients

10:40

who are not treated well, those are the ones that may need

10:42

to be, have neurosurgery.

10:44

So, and again, um, one drug might not work,

10:48

two drugs if they don't work.

10:50

A third drug only increases the chance, um,

10:53

for a seizure free outcome by 5%.

10:54

So really, brain surgery is indicated in a significant

10:57

number of people, unfortunately.

10:59

But in order to have, uh, surgical success, you need

11:02

to find the trigger zone,

11:04

and also when they're operating, do not harm the patient.

11:08

So that's what image is good about.

11:09

So, um, as you know, it's, uh, difficult

11:13

to sometimes find the cause

11:16

of the epilepsy looking at an MRI.

11:18

So here's a patient with, uh, uh, um, epilepsy.

11:22

The, um, EEG, again, cannot localize very accurately.

11:26

So they say somewhere in the left side of the head,

11:28

there's a, uh, epileptic focus.

11:30

So you're looking at a scan like this,

11:32

you know, where is it?

11:34

But what you do though, is we, we could, doing MEG,

11:37

just having a patient, uh, light on the scanner,

11:39

we see a very tight cluster of spikes here that's, um,

11:44

posterior to the insula.

11:45

So, so the insula criteria look posterior to the insula.

11:48

That's where you have to look. So now

11:50

poster the insula, there it is.

11:54

And there it is, surgically confirmed.

11:57

Small focal cortical dysplasia. That's very hard to pick up.

12:01

That's the value of MEG. It tells you where to look.

12:03

Here's another patient. This is a young patient,

12:05

famous, or in San Diego.

12:06

She went to multiple institutions.

12:08

They couldn't really find a lesion.

12:10

They came from MEG,

12:11

and there's a very tight cluster right here.

12:14

Okay, well, then you look again,

12:19

here's where the spikes came from.

12:21

After they tell you where to look, it's obvious, right?

12:24

So it's like the game. Where's Waldo.

12:25

Once somebody shows you

12:26

where Waldo is, it's easy to see 'em.

12:28

But if you're looking blindly, it's tough.

12:30

So, you know, believe it

12:31

or not, this was missed at multiple institutions,

12:34

but here is, um, obviously a, uh,

12:38

focal cortical dysplasia compared to the normal side.

12:41

Um, the surgeon's very confident, instead

12:44

of just sucking this out like they usually do, they,

12:45

they resected on block,

12:46

and sure enough, path proven cortical dysplasia patient is

12:50

doing extremely well, essentially seizure free.

12:54

Here's another example of how the utility

12:55

of MEG this a patient, um, at a major academic center

12:59

who was in a car accident.

13:01

So we, you can see there's ence inflammation in the brain.

13:03

So their question was, well, if there's only one fo, okay,

13:06

and the e the EEG showed, you know,

13:09

activity coming from the, uh, right side of the brain.

13:12

So, you know, their question is, well,

13:14

if there's only one type focus, one small focus

13:18

of activity, we will operate.

13:20

But if you're showing, if you show me, um,

13:22

that there's a lot of, uh, distributed sources,

13:25

we're not gonna just take away huge chunks of brain.

13:28

So we just did the MEG,

13:29

and we can show there's multiple

13:31

generators of spike activity.

13:33

And so they said, great, we don't have

13:35

to do any more, uh, surgical mapping.

13:36

We're not doing anything else. We know there's

13:38

activity coming from all these places.

13:40

So that ends the workup. We, we can't do surgery.

13:42

So that avoided, you know, more invasive, um, uh, uh,

13:46

experiments that had to be done in this patient.

13:49

Um, we can confirm, uh, with, um, Mr.

13:53

Spectroscopy that where the spikes are really,

13:55

is there's something happening.

13:57

So this patient, although there was, um, uh,

14:00

left mesial temporal sclerosis here,

14:02

the spikes actually localized the neocortex lateral to that.

14:05

And that commonly happens that we see sp some, uh,

14:09

spikes from here, but a lot

14:10

of 'em come from the neocortex lateral to it.

14:13

Uh, when you do, um, spectroscopy, you actually confirm

14:16

that, um, the, uh, where the spikes are

14:21

actually, the na is decreased compared to,

14:24

uh, the normal side.

14:26

So there actually is biochemically something going on on the

14:29

other side confirming this, this,

14:33

and again, we can do the same thing

14:34

with frontal lobe epilepsy, that where the spikes are,

14:38

the na is decreased compared to normal.

14:40

Okay, I'm gonna just start, I'll give you, uh, some cases.

14:43

These are just very typical cases that I just, uh,

14:46

pulled from our, our files.

14:48

Uh, 3-year-old female known mesial temporal sclerosis.

14:52

So, um, they wanted to just, uh, confirm

14:56

where the epilepsy foci are from,

14:59

and also do some mapping of, uh, language, uh, which, which,

15:03

which side language is on.

15:05

So when you look at the

15:06

MEG signal, it's, it looks like this.

15:07

There's 306 channels.

15:09

So we break it up into, um, these eight categories,

15:14

these four lobes on each side.

15:16

And then, uh, each one of these is a summation of like, uh,

15:19

13, um, uh, traces like this.

15:21

So right away, you can see here,

15:23

left temporal lobe, there's activity.

15:24

So we can open it up and we can see there's activity in the

15:27

left temporal lobe and a concordant

15:29

with the EEG, which you do simultaneously.

15:32

Okay? And so you can actually, um, uh, here's the signal.

15:37

You can see there's, uh, spikes here.

15:39

So mathematically you just say, where is that spike?

15:42

Mathematically the best fit dipole?

15:44

And boom, I get one spike here.

15:46

And then you look again,

15:47

you see another spike, and you start localizing.

15:49

So I get a very tight cluster right here,

15:51

and it goes basically to the hippocampus

15:53

and the adjoining per hippocampal vis.

15:55

So, um,

15:58

this confirms the obvious left mesial temporal sclerosis,

16:00

we actually don't do it that much

16:02

because as you know, MTS is pretty well solved problem.

16:06

So we don't usually,

16:08

are not usually asked to do it for this.

16:09

'cause they know the answer. Once it's the MTS,

16:11

the EEG shows, left temporal spikes.

16:13

You really don't need to do this.

16:14

But again, when you do it, it's very gratifying

16:16

that we can confirm that.

16:18

Okay? Then, uh, language, what we do is we put, um, uh, uh,

16:22

one word in the right ear earphone, one in the left ear,

16:25

and ask the patient, are these semantically related?

16:27

So their thinking language.

16:28

And so we measure, um, between like 300

16:31

and 800 milliseconds.

16:33

We, we measure over the right

16:35

and left side of the brain, which one is more activation.

16:37

And so we can see here that, um,

16:40

there's much more activity on

16:41

the left hemisphere than the right.

16:43

And so this patient is left hemisphere dominant.

16:46

'cause we're actually seeing the function of the brain, um,

16:49

during the time that they should be interpreting language.

16:51

So we can see them when they hear the language

16:53

that's the auditory, and then hear the word that's auditory.

16:56

But here's where language happens. Okay?

16:58

So then we can actually then map out, you know, axone,

17:02

coronal where the spikes are.

17:04

And then also we can map out the auditory cortex.

17:06

We can, um, here's the median nerve.

17:09

We can also map, uh, brokerage of orage here.

17:11

We just, uh, did just the, uh, hemispheric localization.

17:14

Okay? Here's a patient with a, um, known, uh, occipital,

17:19

cortical dysplasia bilaterally, right?

17:21

Worse than left, um, seizures worsening.

17:25

The EEGS shows something

17:27

that's happening in the back of your head.

17:28

We don't know where it is. So with MEG, we can localize it.

17:33

So again, we can see, look for right occipital lobe,

17:35

left occipital lobe, we see activity, um,

17:38

but spike by spike, we can, um, see exactly

17:42

where in the brain it's happening

17:42

to within a few millimeters.

17:44

So again, it looks like it's really

17:46

coming from the right side.

17:47

And again, spike, after spike, after spike, we localize it,

17:50

and they all go to the, um,

17:52

cortical dysplasia here on the right side.

17:54

Even there's some, uh, um, cort dysplasia on the left side.

17:58

All of the spikes are coming from this side.

18:00

So that really helps the surgeon to know where to go instead

18:03

of saying E EEG somewhere in the back of your head.

18:05

Now we know it really localized exactly

18:06

to the dysplasia right here.

18:09

Um, now, as you know, recently, again,

18:14

we have the, uh, the steroid EEG,

18:17

where we put the electrodes in.

18:18

So, um, everything is confirmed

18:21

before they actually start cutting tissue.

18:22

They confirm it before they'd have to take off the skull,

18:25

put on, uh, grids, um, and so forth.

18:28

Now, they can just put in these, uh, ster, EEG needles.

18:31

But this really is helpful to guide it.

18:33

So they're gonna put a lot right here.

18:35

They're gonna put some right here

18:36

where the other cortical dysplasia is, um, and so forth.

18:39

But they can immediately get, save a lot more time

18:42

and not have to put in as many needles knowing where,

18:45

where the, uh, epileptic zone really is.

18:48

Okay. And again, we give them a, a three dimensional map

18:51

of the different planes, just mapping out exactly

18:53

where the spikes are.

18:54

And this really helps the surgeon, uh,

18:56

guides their placement of the stereo EEG needles.

18:58

And again, we measure them median, uh, nerve.

19:01

And it always goes exactly like it show the

19:03

posterior bank of the central sock.

19:06

Okay? Um, and, uh, okay,

19:09

and then this one, they wanted to know, um, where, uh,

19:14

let's see, this one they wanted to do, uh,

19:17

brokers area and so forth.

19:19

And so actually, is this one with, uh, yeah,

19:23

this is a, sorry.

19:25

This is a patient with tube sclerosis.

19:26

And, uh, you know, you can see the tubers right here.

19:29

So this is very useful in, in patients

19:31

because there's a lot of tubers, right?

19:33

And tube sclerosis. So which one's one causing seizures?

19:36

This is really great for it

19:37

because non-invasively, I could tell you which one

19:39

of these is, is causing the, uh, seizures.

19:42

So you can see there's a tuber here.

19:44

Um, this was one we thought was the most prominent,

19:46

so we were not surprised here.

19:48

But then we saw, um, some activity in right here,

19:52

and it's like, well, there isn't even a tuber here.

19:53

But then, now that we know where Waldo is, we look

19:55

and look, there actually is, if you compare this,

19:58

there's a big tuber right here that's sort of diffused,

20:00

but not as quite as the other ones.

20:02

So this act, this tuber, actually is causing, um, uh, some,

20:07

uh, spike activity as well as the obvious one, which we saw.

20:11

So again, um, it's very helpful.

20:13

And at this time, we also can map out, uh, broke his area,

20:16

um, and, uh, Wernicke.

20:18

So this guy, this person had bilateral bro, his area.

20:21

So we see where that is. We see

20:23

where Wernicke is bilateral wees.

20:24

In this case, we saw auditory cortex.

20:26

So we have the functional map,

20:28

but we also, um, see where the spikes are, um,

20:31

and localized, uh, to where that is.

20:33

With respect to the tubers,

20:35

most time there's only one tuber that's active.

20:38

Um, this case happened to be two,

20:39

and again, it helped us find a

20:41

tuber that was not so obvious.

20:42

So very helpful for patient's tube sclerosis.

20:45

So again, um, MEG, uh, just, uh, summarizing again,

20:49

completely noninvasive, uh, localizes the trigger zones.

20:53

Um, and it, it really is helpful to, uh, uh,

20:57

reduce your invasive studies, optimize the accuracy,

21:00

and it is, uh, clinically accepted.

21:03

Um, MRI safe for all patients.

21:06

All these things that we do are pretty

21:07

safe, but there's radiation.

21:08

Um, um, you have to be careful what, uh, patient safety

21:11

and the MRI scanner, MEG just is not an issue.

21:15

The main cost for MEG, you can, it costs about the same

21:17

as an MRI scanner, but the main cost is personnel

21:20

and the expertise to, uh, do the, do the study.

21:22

The main disadvantages now of MEG,

21:25

because our disadvantages for the number of studies

21:29

that we do, it's very expensive.

21:32

You know, doing an MRI scanner, doing, you know,

21:34

35, 50 patients a day, you know, you can pay

21:37

for it here, we only do a few a day.

21:39

So it's really largely used for research as well.

21:42

Now, the fact that it can diagnose concussions,

21:46

which is another talk, uh, non-invasively

21:50

to 85% sensitivity, much higher than, uh, you know, MRI,

21:55

uh, that could be, uh, a killer app where a lot

21:58

of places might get MEG.

22:00

Um, another disadvantage signal is weak.

22:03

So we, you know, we have to have good shielding, have

22:06

to have a, a dedicated site where with those big,

22:08

a big screen room, and right now there's only about 30

22:11

sites in the United States.

22:13

So with so few sites, it's hard to get acceptance.

22:17

Also, I, I sort of glossed over this,

22:19

but when, um, we're solving for

22:21

where in the brain the activation is happening,

22:24

it's actually non-trivial.

22:26

Um, when if you have distributed sources, so like, it's easy

22:30

for epilepsy, it's very strong,

22:32

single point source, uh, things from media nerve.

22:34

But if I'm looking for things, you know, you know,

22:37

other activity like language, it is more difficult to do.

22:40

But, you know, there's,

22:41

there's some very good mathematical models, um,

22:43

including, uh, Dr.

22:45

MW at our site, uh, is one of the world leaders at creating,

22:49

uh, mathematical models to, uh, solve the, uh,

22:52

where the sources are in the brain from the measured, uh,

22:55

um, megs, uh, uh, signals.

22:59

So now, um, I'm just gonna show some cases, um, some

23:03

of which we'll use Megs to show how it's a problem solving.

23:06

Um, um, I'm giving credit to John Hess link.

23:09

He was the chief at the UCSD, uh,

23:11

before I became the chief of, of UCSD neuroradiology.

23:14

And he, uh, uh, contributed a lot of the cases to this,

23:17

and he's, uh, uh, still a good friend and mentor to me.

23:21

Okay. So, um, epilepsy, you know, the most common cause

23:26

of temporal epilepsy is hippocampal sclerosis.

23:29

Um, there's, uh, you know, uh, you can get vascular lesions

23:33

of caves, and much just examples of these, um, hematomas

23:37

and, um, other cortical dysplasias.

23:40

Um, any kind of time you have, uh, injured brain,

23:43

whether from trauma or ischemia, they can, uh, seize.

23:48

So ence, brain seizes,

23:51

we're just insert some different examples.

23:54

Talking about MRI now, um, it's really important.

23:58

You should do at least three

23:59

tests, MR MRI if at all possible.

24:01

You know, I think everybody who has MRI has a 1.5,

24:04

but I think more, most places probably have three Tesla,

24:07

but you really should do a three Tesla.

24:09

When you compare the same patient on a 1.5

24:11

and three Tesla, you just see so much more in three Tesla.

24:14

Um, now of course, if you have seven Tesla,

24:17

even better still, but that's,

24:18

that's quite, uh, rare at this time.

24:20

Um, so you want to do, um,

24:24

your workhorse is going to be a 3D, uh,

24:29

uh, F-S-P-G-R or MP rage,

24:31

but that way I can get a really good look

24:33

and reformat in the three dimensions of the, uh,

24:37

the cortex looking for chordal heterotopia.

24:39

But other sequences you have to do, um, you have

24:42

to do coronal, um, T two and flare per hippocampus.

24:47

You want ale flare,

24:48

especially the 3D sagittal gets better conspicuity than just

24:51

the, uh, planar, also thinner slices,

24:53

but the lesions show up better as well.

24:55

You want, um, standard T two weighted images,

24:57

and you want gradient echo, um, succinctly weighted images

25:01

to look for blood products, which also can, uh,

25:03

cause be irritant and cause uh, seizures.

25:06

Um, you wanna give galina probably just once, you know, uh,

25:10

just to make sure there's not something, uh, like some, uh,

25:15

well, for example, like, uh, uh, infection

25:18

or something like that, that would enhance.

25:19

But generally, um, the, the yield is low for ga,

25:23

but you do wanna give it at least once.

25:26

Okay? So then you do the T two 80 coronals.

25:28

This is one of your workhorse sequences.

25:30

Um, you, you, here's the hippocampus sagal planes.

25:34

You wanna go perpen to that.

25:35

And so there, you had a nice look here

25:37

that we can see the hippocampus.

25:38

We can see the dentate gyrus.

25:41

You can see the internal structure

25:42

and look for the symmetry.

25:44

Um, so you wanna do that, uh,

25:47

you can look at the fornix as well.

25:49

Okay. So, um, talking about the most common cause of, uh,

25:54

temporal epilepsy is hippocampal sclerosis,

25:56

mesial temporal sclerosis, um, that shows up

25:59

and you have, you know, significant neuronal depletion, 30

26:01

to 50%, uh, most common prominent,

26:04

as you may know in the ca one ca stands for corn amus,

26:08

amman's horn, uh, ca one region of hippocampal body

26:12

and with loss of inhibitory inner neurons.

26:15

Um, and there is some, uh, correlation of history

26:18

of febrile seizures in childhood.

26:20

And so the findings, um,

26:23

really we get such a good look at the hippocampus now.

26:26

So, and you could do volumetric measurements as well

26:28

to actually quantitatively measure the

26:29

asymmetry between them.

26:31

But the left and right, of course, you know,

26:34

a certain significant, maybe 10% of 'em can be bilateral.

26:36

But, um, you can measure the atrophy

26:39

and also see it by eye comparing left versus right.

26:42

But there, there're programs now that when we use those

26:45

to actually get quantitative measurements of the asymmetry,

26:48

uh, hypertensive and T two

26:49

and flare images, very useful sign.

26:52

It's, uh, essential actually to see this, um,

26:55

this temporal horn dilation and, uh, is not as useful.

27:01

That was not in the days before we had good MR mri.

27:04

So to see the hippocampus, you couldn't see it

27:05

that well on CT in the old days.

27:07

So you look for temporal dilation.

27:08

Now, this is not that useful

27:10

'cause I don't need to look at the horn around it.

27:12

I just look at the hippocampus directly.

27:14

But, you know, you still look at this

27:16

and, uh, sometimes not infrequently, there were actually we

27:20

atrophy of the whole temporal lobe,

27:22

but not just the, uh, hippocampus.

27:24

So just show an example.

27:26

Um, the, this is very common,

27:28

and I think you all are well aware of this.

27:30

Here we have a, a, the 20 or bone with epilepsy.

27:33

You can see the, the right hippocampus is brighter.

27:36

I still see some, uh, the internal structure,

27:38

but better than I see in some cases.

27:40

But it's, it's, it's bright.

27:42

So this is a sign of, um, uh, hippocampal sclerosis,

27:46

and you can even see it on the axles,

27:48

but you know, it's tough to see on the

27:49

axxis compared to therons.

27:50

So you always wanna look here

27:52

and you can see, oh, the temporal one's a little bit bigger

27:53

here, but you, the theron, the coronals is, is where you,

27:56

you is the money, okay?

27:58

And see on T two, same thing.

28:00

You can see the, it's right on T two as well as flare. Okay?

28:04

Here's another case.

28:05

This is a, a young woman with, um, you know, epilepsy.

28:09

I'm just pointing out this thing right here.

28:12

This is not pathology, okay?

28:15

So you see this routinely as well.

28:17

This is a choroidal fissure cyst.

28:19

This is a, uh, benign incidental finding.

28:21

Here's the choroidal fissure.

28:23

And, uh, this, uh, you know, is, uh, a common finding.

28:28

It is not related to, uh, seizures.

28:32

Okay? Uh, another patient with seizures, 30 or 3-year-old.

28:36

And, um, again, you can look at the, uh, hippocampus

28:40

on sagittal images and not just on the axles and coronal.

28:44

So here on the left side,

28:45

we can see a nice plump hippocampus here.

28:48

See how thin it is on the right side.

28:50

And it's also right here.

28:52

So use, use all your planes,

28:54

but we get a good look also on the sagittal view.

28:56

And of course, it's confirmed normal left hippocampus

28:59

and a bright, uh, small, uh, uh, right hippocampus.

29:04

And it's shrunken here.

29:06

And it's same thing shows up on the, uh, flare. Okay?

29:10

Another case, just giving a lot of cases of,

29:11

um, meso temporal sclerosis.

29:13

Alright, now here's another case.

29:16

20-year-old female presents with fever and seizures. Okay?

29:20

So you look at this, well, I mean, yeah,

29:23

the hipc camps is bright,

29:24

but there's also is involved in amygdala.

29:27

And, uh, this is the,

29:29

does not look like meso temporal sclerosis.

29:31

It looks like there's a sort

29:33

of a more diffuse process going on here, especially fever.

29:37

You've got to think of, uh, a herpes simplex.

29:41

'cause if it goes untreated, it's devastating.

29:43

But if you, uh, treat it right away,

29:45

you can really stop the patient's progression of disease.

29:48

So this is an example of, uh, herpes simple.

29:51

They give, when you give contrast,

29:53

you can see it's a little swollen here,

29:54

slight enhancement, that's typical.

29:55

You get this slight enhancement of necrotic regions,

29:58

and it's not an infarct.

30:00

There's no restricted diffusion. So this way, this is, um,

30:03

um, meso temporal sclerosis, uh, uh, uh, herpes simplex, um,

30:08

treated, you know, seven months later,

30:11

you can see there's now sort of progressed through.

30:13

It's a smaller and brighter hippocampus.

30:16

So it's, um, herpes simplex, which is now, uh,

30:20

with a end result of some, uh, uh, hippo sclerosis.

30:28

Okay? Here's a, uh, child with seizures.

30:32

So we see this thing right on T two,

30:36

could be some li some cystic component here,

30:39

but it's, um, cystic component

30:42

would not be this bright on flares.

30:44

This is, uh, maybe not a bunch of cysts.

30:48

And again, it's almost looks,

30:50

almost could be cyst like here, but enhanced with contrast.

30:54

So the first thing you think

30:55

of a child like this, uh, dnet, right?

30:58

Dysplastic neuro epithelial tumor could

31:00

look like this, actually.

31:02

Um, and that's common.

31:04

But in this case, because it's enhancing like this,

31:07

it's less typical for that.

31:08

It would be. So this comes up a ganglion cell tumor.

31:11

But if I, but just think if I see little cystic lesions

31:13

and temporal lobe trial seizures, uh, DN ts most likely.

31:17

And, uh, you will see one, you will see D nets, okay?

31:21

Hemorrhagic lesions that can cause, um, seizures.

31:24

Well, yeah, when you see a bleed like this, you, it's harder

31:27

or hard to tell the cause of the bleed here.

31:29

But it, uh, turns out the, uh, cavernous angio

31:32

or, uh, cavernous malformation that bled.

31:36

You can see the hematoma right here.

31:38

Um, even if they don't bleed, you can tell the typical, uh,

31:43

in internal popcorn mulberry appearance, um,

31:46

and hemo, citrine rim, those can't cause seizures.

31:49

And this one happened to cause bleeding, which is even more

31:51

of a irritant.

31:54

Okay? Here's another cause of seizures.

31:57

Patient with, uh, the, uh, um,

32:02

can't read the top of this thing,

32:03

but, you know, uh, lightheadedness.

32:05

Oh, visual disturbance.

32:06

So, um, this is not a diagnostic dilemma,

32:11

but we can see this weds shaped, uh, lesion right here,

32:14

but it's also bright right here.

32:17

And, uh, we, the diffusion weighted images, we can see

32:21

that there's an infarct in the occipital lobe

32:24

and here in the hippocampus.

32:26

So, um, this is the posterior body hippocampus is supplied

32:30

by the anac choroidal, but also poster choroidal as well.

32:33

So this, um, this is an infarct of the hippocampus, which,

32:37

uh, is causing, uh, seizures as well

32:40

as the hemianopsia.

32:45

Okay? Again, important to know this, um,

32:51

seizures, increasing confusion in lethargy.

32:54

Um, you know,

32:58

you don't wanna miss this thing here.

32:59

There's this, so there's T two prolongation, um, here

33:03

around the insula.

33:06

An cortex or perm, cingulate sulcus.

33:10

It likes the cingulate too, right?

33:12

As well as the hippo campi.

33:14

So, um, hippo campi involves, well, so this is,

33:18

again, a herpes simplex.

33:20

You must think of this because you must give

33:22

the acyclovir right away.

33:24

Um, and again, very little enhancement.

33:26

Um, but this is, uh, very typical

33:30

where it can be bilateral,

33:31

but not, um, symmetric, hippocampal involvement,

33:35

other parts of temporal lobe.

33:36

But it likes the cingulate, you know,

33:37

and it likes the insula.

33:40

So as soon as you see that, you immediately, uh,

33:42

call the pharmacy to drop acyclovir, talk to the, the id,

33:46

uh, the infectious disease team and treat this patient.

33:51

Okay? A 20-year-old, a pregnant woman,

33:53

now have several seizures daily.

33:55

So when you have a pregnant woman who has seizures, um,

33:58

you think of eclampsia, yes, she had high blood pressure.

34:00

She, this is a patient that eclampsia

34:02

with constant seizures.

34:04

And so we have this swelling of the, uh, brain here

34:08

of the, uh, temporal lobe.

34:10

And, uh, this is, um, uh,

34:16

uh, edema related to status epilepticus.

34:21

So it, it does happen.

34:23

And, and you will see this, somebody

34:25

who has status epilepticus a lot of seizures,

34:27

they will get edema in their, um, of the, uh, temporal lobe

34:31

or whatever part is seizing.

34:32

So you've just gotta control the, control,

34:34

the blood pressure and control the seizures.

34:37

And this thing, uh, will, will regress, okay?

34:41

In infectious process,

34:42

this is living in southern California, uh,

34:45

or southwest United States.

34:47

We see this cavity in the brain right here, um,

34:52

sort of walled off cavity right here.

34:54

I'm gonna show you a picture

34:55

that will blow your minds in a few seconds.

34:58

But, uh, so when you see this thing right here, um,

35:02

Southern California or Mexico, you have to think

35:04

of neurotics.

35:05

There can be other parasites and so forth that do this,

35:07

but, um, it's neurotics is so common, you know,

35:12

that the, the San Diego, we see it

35:14

'cause we're, we're closer to

35:15

Mexican than we are to Los Angeles.

35:18

And so people come in car accidents,

35:20

we see calcifications in their brain, it's says, oh,

35:21

another case of neuro psychosis.

35:23

We will see this at least weekly.

35:25

Um, and then, uh, you can see, oh, yeah, look, I can sort

35:29

of see on this one flare.

35:30

I can sort of see this thing. Okay?

35:33

That's the little, the little worm, okay? Inside there.

35:37

But you can see it much better with another sequence. Okay?

35:40

So there's a larvae right there.

35:41

So this is neuros psychosis, uh, I guess, um,

35:45

then you treat it, and then you get an

35:47

inflammatory reaction.

35:49

And, uh, so in the inflammatory reaction, the,

35:51

the thing is dying, but then we get this edema, okay?

35:55

So here's what the way you want to image this.

35:58

You wanna do fiesta, uh, or the fiesta sequence heavily.

36:02

T two weighted. The same thing you would do is like

36:04

for the iacs.

36:05

Um, one of my fellows, uh, was John Butman,

36:08

who's now the chief of, um, neuro radial, NIH.

36:10

And he showed me this. We do this all the time now.

36:12

And it just remarkable

36:14

because the thing is, you know, with regular T twos, um, uh,

36:18

edema is bright, juicy worms are bright.

36:21

And so you can't tell a juicy worm from the fluid.

36:24

But on flare, on fiesta, only water is bright, so

36:28

that juicy wet worm is dark.

36:31

So I can see the larvae very clearly compared to fluid.

36:35

And so when you do this, you, you always see the larva.

36:37

So every time there's a neurosurg

36:39

osis, we always do the fiesta.

36:40

And not only do we see the cyst, now we see the, the worm,

36:44

which is W white on T two 'cause it's juicy and wet,

36:47

but it's dark on Fiesta.

36:49

So whenever fiesta is a great sequence, uh, for neurotics,

36:55

now you look at this case, when I look at this,

36:59

RAs Amos Neurotics looks exactly like this.

37:01

This is exactly an appearance of RAs, Amos neuro cystics,

37:06

but the history is chronic immunosuppression

37:08

with steroids and methotrexate.

37:10

So this is a very typical picture

37:15

of tb.

37:17

And so those two things are identical.

37:19

So you have to separate out by the history.

37:23

Um, so, uh, TB looks exactly like this.

37:27

You can also, as you know, get tuberculosis.

37:29

But when you see this thing,

37:30

there's two things that look like this.

37:32

It's either SMOs, neurotics,

37:34

or tuberculosis, especially in immunocompromised.

37:38

And so you need to get history.

37:40

You know, obviously CSF will help you.

37:42

Um, if you're on the East coast, you might never see,

37:45

um, neurotics.

37:47

And so you'll get the diagnosis, right?

37:48

If you're in the California,

37:50

or you know, southwest, you see so much neuros psychosis,

37:53

you don't even think of tb.

37:55

But in fact, this is tb. And, and we, we know this, okay?

37:58

You can also get developmental lesions, um, you know,

38:01

lissencephaly, RIA, RIA case, poly micro gyre, um,

38:06

gray matter, heterotopia, anencephaly.

38:09

Um, you can get, um, microcephaly

38:12

or mega cephalic can cause seizures as well.

38:15

And the tube well known, uh,

38:17

OSIS tube sclerosis and Sturge Weber.

38:20

So here's an example of a chy, gyre and lissencephaly.

38:24

There's the very poor gyration

38:26

and just thick, uh, uh, thick cortex.

38:30

Here's another case that just came across.

38:32

I'm gonna do an MEG on this, um, in a couple weeks.

38:35

Uh, so I always get the MRI first.

38:37

So, uh, we're gonna try and do MEG

38:39

and see where the spikes are coming from here.

38:41

But you can see there's poor ation

38:43

or abnormal ation right here,

38:44

but also just this thick band of, um, of very thick cortex.

38:50

So this is, uh, looks like it's P area, abnormal location.

38:54

So we'll see what the MEG shows. Okay?

38:58

This is something I want you guys to know.

39:00

This is commonly missed. This case was missed.

39:04

In fact, I just turned in the MEG report yesterday from this

39:07

patient, um, where it was seen at multiple institutions

39:11

and they said, um, you know, normal, normal MRI.

39:16

And, uh, but you ha always, when you ever do an epilepsy,

39:19

you have to look at the temporal, uh, pulse

39:23

and look carefully at the gray matter, uh,

39:25

the gray white junction and the white matter in particular,

39:28

and make sure the white matter is equally

39:30

dark on both sides.

39:33

So here's a T two,

39:34

and it's the kind of thing, if you look for it, it's

39:38

for the per party of a template.

39:39

When I do my epilepsy conference every week, you know, I go

39:43

through the hippocampal, but I always

39:44

look at the temporal poles.

39:46

You have to do this because this is, um,

39:48

it's a known entity, but it's just, it is missed.

39:51

So clearly there's something wrong with this.

39:54

The, uh, white matter is just too bright.

39:56

And there's also some blurring of the, uh,

39:58

gray white junction compared

40:00

to the other side on flair confirms it, right?

40:02

And it, it's kind of subtle, right?

40:04

But, you know, this is the normal darkness

40:07

of the white matter, and it's just not as dark.

40:10

This is definitely abnormal, okay?

40:13

And again, there's blurring of the, uh, there's blurring

40:16

of the junction as well, okay?

40:18

And the axials say, same thing when you look at the axials

40:21

black, the white matter is, is dark,

40:23

and here's just not as dark, okay?

40:26

And I go, another, another slice.

40:27

It can be this subtle, right?

40:29

So here you look, it's dark and it's, it's there,

40:33

but it's just not as dark.

40:37

There's the pet. So pretty striking, right?

40:42

It's clearly abnormal. So this is a focal cortis.

40:46

It's well known. The temporal poles in particular,

40:50

commonly the literature says it's associated

40:53

with mesial temporal sclerosis.

40:54

In my experience, I see this

40:56

often without mesial temporal sclerosis on this, on

40:59

that ipsilateral side.

41:01

Okay? So we did MEG. Okay?

41:05

Again, we look at this, uh, we get right temporal spikes.

41:09

You can see the right temporal spikes here,

41:11

and nothing on the left side.

41:12

It correlates to the EEG.

41:15

And so, again, right side, I'm seeing the spike activity,

41:18

nothing in the rest of the brain,

41:19

so it's gotta be somewhere there.

41:20

But we, we say, okay, assign one DPO to account for this,

41:24

this, uh, uh, signals across all the sensors,

41:28

and you do that, bam.

41:29

I get one dipo right here, and we keep doing it,

41:32

and we get dipple after dipple.

41:34

So I got a whole cluster of spikes right here, um,

41:37

involving the, uh, posterior aspect of temporal pole

41:40

and, uh, some of the hippocampus as well.

41:42

So it's sort of at the posterior border of where the, um,

41:47

uh, fibro focal cord dysplasia is.

41:50

A lot of times the epileptic spikes won't go

41:53

through the whole region,

41:54

but it'll just go to the border of it,

41:56

or it'll go to the, if I have a lesion, like in the,

41:58

like I said in the hippocampus, it'll go

42:00

to the neocortex, lateral to it.

42:02

So this is sort of at the posterior border of where the, um,

42:06

uh, focal cord dysplasias a case we

42:09

did just a couple months ago.

42:11

The whole thing spikes all over here.

42:14

The whole thing did have spikes.

42:15

This one happens to be at the poster border of it.

42:18

But it's just really helpful because again,

42:20

if you didn't see the lesion, um, on the MRI,

42:23

which it wasn't seen, you go back to this

42:26

and then you see right away to look at that temporal pole

42:29

and, uh, you, and you'll pick it up right away.

42:31

The hippocampus in this case was normal.

42:34

And again, we can, uh, show in x, y

42:36

and, uh, sax andron all where all the spikes are.

42:40

So this will really guide where the surgeon goes.

42:42

Um, she'll put a lot of, um, steroid eg needles here.

42:45

She'll also do other places there.

42:46

But, um, this is going to really help the surgeon.

42:49

Um, and of course, we can measure the median nerve activity

42:52

where you can measure the auditory cortex.

42:58

Here's another lesion that you need to know about

43:00

that can cause, um, epilepsy, this, uh,

43:02

encephalocele, right?

43:05

So, um, uh,

43:08

just here's one cut that's next cut.

43:11

And ence seals, especially in temporal lobes

43:13

can cause uh, um, epilepsy.

43:16

And this one here is, uh, on the axial view.

43:20

You have to find these, you in addition,

43:21

look at temporal pole, look on those coronals look

43:24

for ence seals, especially in temporal lobe.

43:26

And pet scan was great. Here.

43:28

It shows exactly the, uh, the absent,

43:33

uh, uptake corresponding to the, uh, epilep xic Joan.

43:37

Now this case we did MEG.

43:39

And I'll be honest, MEG didn't find any spikes.

43:41

That happens. If you have somebody who has epilepsy,

43:42

you put 'em in E eeg, sometimes you just don't see spikes.

43:45

So the MEG this case didn't see spikes,

43:47

but, you know, um, the pet

43:50

that was confirmatory in this particular case.

43:52

Now, here's another case. Here's a focal

43:54

cortical dysplasia right here.

43:55

Now, you know that the, um, neurons go from the ventricles,

43:59

the radio, uh, glial, um, along the radio gl

44:02

and they go laterally.

44:05

I found that when I look,

44:07

when I see this focal cor displease,

44:09

I often don't see abnormality here in the neocortex.

44:12

I just, just, I just can't see something that looks like it.

44:14

I can see a little patch right here coming to it.

44:17

Um, anyways, um, here's the one case

44:20

where the pet was negative.

44:21

PET didn't see anything, but we do the MEG

44:25

and, uh, there's no cordial dysplasia.

44:28

The pet was negative. So here's the opposite.

44:30

And here we see, we look for the, um, uh, activity

44:34

and we can see, especially in the right frontal lobe.

44:36

So there's activity in the temporal lobe,

44:37

but especially in the frontal lobe.

44:39

Okay? So again, we can localize where it is with the MEG

44:42

and we see this, all these spikes,

44:44

and they're going to the neocortex lateral

44:46

to the focal corti dysplasia.

44:48

So again, maybe there's some activity here, we don't see it,

44:51

but this really got, helps guide the, the neurosurgeon.

44:54

'cause believe me, they're gonna put a lot of, uh, uh,

44:57

sterile EEG needles here, as well as here to help them know

45:00

what they're gonna do if they're gonna do laser

45:02

interstitial, uh, thermal therapy, so et cetera.

45:05

But this really helpful to see all the spikes are coming

45:07

from here, whereas the EEG said somewhere in the,

45:11

the right frontal lobe, now we can localize it.

45:14

And again, the pet was negative, but MEG was helpful.

45:17

And again, we can show where all the spikes are.

45:19

Okay, so, um, here's another, just some examples

45:23

of focal cordal dis uh, uh, uh, uh,

45:26

cordal dysplasia right here.

45:27

You can see this abnormal, um, cortex right here.

45:34

So, again, not a surprise that this patient has,

45:36

uh, epilepsy.

45:41

Now, just another example of, of,

45:43

of what we see all the time.

45:44

Look at the temporal poles.

45:46

And again, we just see right here this blurring.

45:49

The, the white matter is not as dark here.

45:51

And there's blurring of the, um, uh,

45:54

gray white junction right here.

45:56

This, this case, the amygdala's involved. Okay?

46:00

And the pet, um, again,

46:02

shows this is a focal cord dysplasia cause

46:04

of the patient's epilepsy, um, man

46:08

with seizures for many years.

46:10

You got have to look another part of search strategy.

46:12

Look along the ventricles.

46:13

T two I find is pretty good,

46:14

but also the, um, the, the T one weighted, uh,

46:18

thin section MP rage or F-S-P-G-R.

46:21

And you can see right here, obvious nodules lining the tire.

46:25

Left lateral ventricle right here.

46:27

See, there's nice white matter lining.

46:29

This one, just lots of nodules right here.

46:32

And again, see the nodules right here.

46:33

So again, this is, you know, cortical heterotopia

46:36

and that's the cause of patient seizures.

46:38

Now, when we do MEG on these, a lot

46:41

of times sometimes we see that this thing have activity,

46:44

but a lot of times we don't.

46:45

And it really is a, the neocortex lateral

46:48

to it, amazingly enough.

46:50

Uh, and you can also see maybe here's a trans

46:53

mantle sign here, who knows?

46:54

But yeah, so that's helpful.

46:56

'cause the activity is not always in these nodules.

46:58

In fact, usually it's more likely

47:00

to be out here than here when we do the MEG.

47:02

Uh, this patient, uh, uh, not have NMEG.

47:05

Uh, lastly, patient with seizures.

47:07

Um, so this obvious, you know, um, tubes sclerosis,

47:12

um, with these tubers, you know, through the,

47:15

the calcifications right here along the ventricles.

47:18

And again, like I said, meg's really helpful

47:20

because all these tubers, generally, there's only one

47:24

that will cau be causing seizures.

47:26

And so that's where we do it.

47:27

We find that one, and that's when the one, the,

47:29

the surgeons will concentrate on.

47:31

And I think the last slide here, Sturge Weber,

47:33

is also a cause of seizures.

47:35

We see the, uh, the, uh, uh, peel angio mitosis

47:40

as you know, the, uh, OID plexus on

47:42

that side can also be large and enlarged.

47:45

And, uh, you know, the, uh, court Weinstein

47:48

and all the other things go with it.

47:49

Anyways, so here was a tour through, um, uh,

47:54

first talking about MEG mag ence holography

47:57

and how it's helpful, especially for epilepsy, showing many,

48:00

many cases of how MEG was useful in epilepsy.

48:03

And then just showing, uh, for people don't have an MEG, um,

48:07

many is how you, how MRI is very helpful in diagnosing it.

48:12

Uh, by the way, even if you don't have

48:13

an MEG, you don't need to.

48:16

You just, uh, refer to the, the centers you know,

48:18

that are near you.

48:20

So there's only two in California.

48:22

Uh, well, Kaiser has one in Northern California.

48:24

Now there's one at UCSF and one at UCSD.

48:27

So the hospitals around me, you know, um,

48:30

the private hospitals, sharp

48:32

and Scripps, um, the VA where I work Navy, uh, UCLA

48:36

and Irvine send me cases USC.

48:38

So you can, uh, uh, refer

48:40

to your MEG center 'cause it will be helpful.

48:43

Um, and so I'm happy to, uh, uh, do MEG on, on my, uh,

48:48

colleagues, you know, around, around Southern California.

48:52

Okay? So, uh, I'll stop sharing and, uh,

48:56

or if there's any questions, I'll be happy to, uh,

48:59

see if I can help answer your questions.

49:01

Yeah. Thank you so much, Dr. Lee.

49:03

At this time, we will now open the floor

49:05

for any questions from our audience,

49:06

and you may submit your questions

49:08

through the q and a feature.

49:10

It looks like you've got three in there right now.

49:13

Dr. Lee, if you see those. Yes.

49:15

Okay. Yeah, Mary Solomon, uh, squid units.

49:17

Fortunately it's not $10 million each.

49:20

It's about size of an MRI.

49:22

Um, and also the helium is

49:24

recycling is now contained in all of them.

49:25

So you don't have to pay for helium.

49:27

Um, either like maybe, uh, refill less than once a year.

49:31

It's more like 3 million, um, uh, with a screen room.

49:36

And that may or may not include a screen room.

49:37

If you have a really expensive

49:38

screen, we might add another million.

49:40

But I think overall we, for like 3 million, you can do it.

49:42

So it's really the cost of an MRI now, um,

49:46

of course it's great to always optimize MRI

49:49

and PET techniques, um, and people are doing that.

49:52

And we, every case that we do, we always do m mri, we do,

49:55

uh, MRI with the different sequences I talked about.

49:58

So we're getting quite good at doing it.

50:00

And we also do pet on every patient,

50:02

but we also do MEG on any

50:04

of the tough patients, which is most of them.

50:06

Um, but there's something around the, the, the bend.

50:10

And that is, there's now going to,

50:14

there are now actually on the market MEG units that,

50:19

um, don't need liquid helium,

50:20

that don't need

50:21

superconducting quantum interference devices.

50:23

They use optically pumped magnetometers.

50:27

And so those are just like a little, um,

50:30

little cartridge about the size of your finger,

50:33

like index finger and inside that's a, um, a laser

50:37

and like a rear, I guess

50:38

like rubidium or something like that.

50:40

And by pumping, do optical pumping, um,

50:44

you can cause transitions by the magnetic fields.

50:48

You can actually measure the magnetic fields.

50:49

It's not as sensitive a squid,

50:50

but it's good enough to actually teach

50:53

the brains magnetic fields.

50:54

Um, and one reason it's, it's so that way

50:57

because the squid helmet is, you know,

50:59

because the liquid helium around the squids,

51:01

you can't get right off top of the brain.

51:02

These little, um, squid sensors,

51:05

you can put it right on and touch your head.

51:07

Okay, touch the, so you actually get closer.

51:10

So the signal is stronger in that way.

51:13

So it's giving equivalent signal noise as the other ones.

51:16

And if it's gonna be, you know, five times,

51:19

10 times cheaper, so then it comes on the order of,

51:23

you know, a million and maybe with something

51:26

for the screen driven, you know, it, it can drop.

51:28

'cause you know, it's just a laser.

51:29

And so the prices coming in, so we, we,

51:32

and up probably most sites are, are working with those now,

51:36

but there are commercial ones too.

51:38

We have, uh, got a bunch of, uh, um, these, uh, ly pumped,

51:42

uh, sensors and we've, uh,

51:43

we're doing some experiments with them as well.

51:46

And, but there will, there are some hole head ones already.

51:49

Um, Birmingham, England has some,

51:52

so the price is coming down.

51:54

And frankly, once that price comes down, guess what

51:58

the squid ones are gonna come down to.

52:01

Um, 'cause you think about it, how much does it cost

52:06

to fabricate?

52:07

Uh, um, uh, uh, uh, a little a sensor.

52:13

You know, they're, they're

52:14

charging what they can get, right?

52:15

But the cost for that, if you think about it, how many,

52:18

you know, much more complicated electronics is in my

52:21

iPhone, right?

52:22

And how many, you know, millions of trans, you know, uh,

52:26

you know, transistors are there.

52:28

This is just one sensor.

52:30

So the price will come down, uh, of the squids as well.

52:33

Anyways, um, our oblique

52:39

better, better than T two in addition to T two flare.

52:44

Oh, yes. Uh, DI absolutely, we, so we, we can,

52:47

we could, we can do this.

52:49

Um, the double inversion recovery, um, actually

52:52

what I think about double inversion recovery, you know, we,

52:55

I was one of the first ones to get, get

52:56

that one G came out with it.

52:57

And, uh, I actually like the flare queue better.

53:01

Um, so, so we can do that.

53:05

There's a sequence called Gator, which is really good too.

53:08

Um, so you can use, yeah, so I, I, I support all

53:11

of those things, but I think that the, the best, some

53:13

of the best sequence, uh, that I is, that, that 3D uh,

53:16

t two flare, um, reformatted, I think

53:19

that's better than the DIR and, uh, in a lot of cases.

53:23

And, uh, and I think that, that, that gator, that's the,

53:26

that ir is also really good too.

53:27

But yeah, all, all the sequencers are great.

53:32

Uh, future growth of using MEG again, um, MEGI think

53:35

as people see it's useful.

53:37

I think it's, it's going to grow.

53:39

But especially once the other techniques, like

53:42

for diagnosing concussions, we're seeing, um, we,

53:47

we can see the slow waves, which is the brain function.

53:50

So, you know, MRI

53:51

or DTI, you're looking at brain structure, we're, we,

53:55

we're actually seeing the brain function.

53:56

So the rhythms are abnormal.

53:57

We've shown in countless publications

53:59

and other sites across the country world are doing it too.

54:02

You, if you just look for the slow waves, um,

54:05

you can diagnose concussions.

54:06

We, we've published 85% sensitivity, um,

54:09

with no false positives.

54:11

Of course, EEG can be done,

54:12

but they're not as sensitive as MEG is.

54:14

And now with these, uh, optical pump things

54:16

where you don't need liquid helium, um, you know, uh,

54:21

the price is gonna come down.

54:22

I think MEG uh, will grow, especially

54:25

with this optical pumped, uh, optical pumped, uh,

54:28

devices research.

54:33

Okay? Yes. Um,

54:37

further reading, there's a lot.

54:39

Uh, well, you know, I, few years ago,

54:43

neuro neuroimaging clinics of North America, I wrote a, uh,

54:47

uh, a whole little volume.

54:48

My colleague Mhu and I

54:50

and other top imaging people across the globe, I got some

54:54

of the top people that my friends to write it for me.

54:56

And so I wrote a, uh, monograph.

54:58

So if you look for neuroimaging clinics of North America,

55:01

that's a nice introduction 'cause it's short

55:04

and it just talks about, you know, what you need to know

55:07

because it's for radiologists.

55:08

So, um, talk about how it works,

55:11

but also how it's used in, you know, epilepsy written by one

55:16

of the top people in the field, neurologists in Cleveland

55:17

Clinic, who does more than anybody else.

55:19

They're both chapter in epilepsy,

55:21

but also, um, you know, for our work on concussions,

55:24

schizophrenia and so forth.

55:26

So that might be a good thing to start.

55:28

But if you just Google, there's tons

55:29

of great literature on MEG Caribbean, I'm not aware

55:34

of the Caribbean, um, having an MEG,

55:37

but I think Florida has some, I mean, if you,

55:39

you can get somebody that states, uh, uh, Alabama, uh,

55:43

Birmingham I think has, has one.

55:45

Um, so, uh, yeah, I just

55:48

come, come to the United States.

55:50

There's, there's again, several in the United States.

55:54

If you wanna come to San Diego

55:55

and have good weather like the

55:56

Caribbean to come to San Diego.

55:58

But wherever you go, I mean,

55:59

they're all, all the places are great.

56:00

UCSF is great, you know, um, Harvard, uh,

56:03

mass General Steve stuff is superb.

56:06

So just come up here.

56:08

We all welcome you, our typical metabolics FG pet.

56:13

Um, yeah, I mean, yes, I think

56:17

we haven't studied ence flights per se,

56:18

but there could well be spiking, like if you do your EEG,

56:22

you might find that we could localize where the spikes are.

56:25

Um, but also, uh, we can see abnormal brain rhythms.

56:30

So again, but EEG is, um, would also be able

56:35

to see the brain rhythms.

56:36

We a little bit more sensitive.

56:37

So, um, you know, I think if you,

56:42

again, if you can, it is probably more

56:44

and more helpful to actually do the, uh, the antibody panel

56:48

to, to, um, to make the diagnosis.

56:51

Even the MRI's normal.

56:53

Um, you would probably see abnormal stuff on p on, on, uh,

56:56

MEG as well as pet.

56:58

But I would say, um, if you,

57:01

the autoimmune panels is the definitive,

57:04

but it'd be interesting

57:06

to M-E-G-I-I bet it at least would show abnormal rhythm.

57:11

Um, yeah, I mean, we can use ME for any disease.

57:16

Um, you can, uh, MEG for disease strokes, we,

57:20

we will definitely see slow waves there.

57:23

Um, on the other hand, I think, uh, something like, uh,

57:26

stroke, it's so well seen with, um, epilepsy, uh, with, uh,

57:31

with, uh, MRI that, uh, you know, there's not that much need

57:35

that I can see for MEG

57:37

because with MRI, we can see it so well

57:41

and, um, you, you know,

57:43

but, uh, yeah, we, we can,

57:45

we do occasionally do em egen stroke patients

57:47

to see the slow waves,

57:49

but, um, it doesn't really add that much,

57:53

most common cause of missing epilepsy.

57:54

And MRI, I think a lot of things I try to point out,

57:59

I think the hippocampus is pretty well seen,

58:02

but what I've seen is the, um, that looking

58:06

for the white matter,

58:08

the abnormal white matter in the temporal poles,

58:10

that's like the case I just showed.

58:12

That was a real case that just, I just did imaging from it.

58:14

So, uh, so that's the most common thing.

58:17

Missing that focal cord dysplasia and temporal poles,

58:20

but also look for ence encephalocele,

58:23

I've seen that missed as well.

58:25

Um, and then there are, it's hard

58:29

to see cord focal cortical, well,

58:31

cortical dysplasias in general are sometimes hard to see.

58:34

And so what I think, uh, you're probably doing this,

58:38

but use other information.

58:40

So you always will have an EEG, you know, a good EEG video,

58:44

EEG, use the pet, then go back

58:47

and look at your MRI get good 3D um, high resolution,

58:53

um, you know, the 3D flare cubed

58:57

and the MP rage and just look again.

59:00

Um, that's how I see things that I, i I missed.

59:02

Like, you know, if um, if I, if I hadn't missed that, um,

59:07

temporal lobe thing only,

59:08

'cause I always looked for it once I saw the MEG abnormal

59:11

there and the pet, you go back

59:12

then, you'll definitely see it then.

59:14

So I think I've given a lot of tips about the places

59:17

to look, but the main thing is just be diligent.

59:20

Knowing this really can change somebody's life.

59:22

If you have make, find a lesion, um, it changes things

59:28

you, yeah, they can live on, you know, stereo, eeg,

59:31

Noer G or a pet.

59:33

But if you can see a lesion on the MRI, the surgeons are

59:35

so grateful and the patient will do, do, do much better.

59:39

So just realize what you're doing, you're doing epilepsy.

59:43

MRI is so important. And, um, just be diligent.

59:47

Use the other information, the EEG especially, and go back

59:51

and you'll find lesions you missed every so often.

59:57

All right. Thank you so much Dr. Lee.

59:59

I think you got through all of 'em.

60:02

Great. So thank you so much

60:04

for sharing your expertise with us today.

60:06

Yeah, I've really, uh, uh, honored to be able to speak

60:10

to this, this audience.

60:11

And, uh, again, um, my, my email address is simple.

60:16

It's RR lee

60:17

rl@ucsd.edu, you know, university

60:21

of California san diego ucsd.edu.

60:25

So please email me.

60:27

Um, also mod can probably make it available, Gil,

60:29

but it's RRL e@ucsd.edu.

60:31

It was on my first slide.

60:33

Please email if you have any questions,

60:34

and I'll be happy to talk to you.

60:35

I can give my, I'll give my cell phone at that time

60:37

and we can talk.

60:39

Awesome. Thank you so much again,

60:41

and thank you so much to everyone

60:42

for participating in our noon conference

60:44

and asking great questions.

60:46

You can access the recording of today's conference

60:48

and all our previous noon conferences

60:50

by creating a free account.

60:52

We'll also email out a link to the replay later today.

60:56

Be sure to join us on Thursday,

60:58

June 19th at 12:00 PM Eastern,

61:00

where Dr. Harris Cohen will deliver a lecture entitled

61:03

Perinatal Neuros Sonography Review,

61:05

and some teaching points you can register for@mrionline.com.

61:10

And follow us on social media

61:11

for updates on future noon conferences.

61:13

Thanks again, and have a great day.