0:00

Isotropic imaging.

0:03

It's key for coronary.

0:06

So what exactly is it?

0:07

The term basically means that the spatial

0:12

resolution is the same in all three

0:17

planes, three planes being X, Y, and Z.

0:23

So spatial resolution is very important for coronary

0:29

because coronary arteries tend to be very small.

0:31

So we want to see the small diameter structures

0:35

and be able to quantify the degree of stenosis.

0:37

There are two types of spatial resolution.

0:39

There's the in-plane.

1:41

Which is the XY plane, which is the lower-lying

1:45

fruit because all you have to do is increase

1:48

the matrix size, narrow the field of view, and

1:51

you can get very high spatial resolution.

1:54

The more difficult part is getting a

1:56

spatial resolution through the Z axis.

1:59

So XY will create the pixel, Z will create

2:01

the voxel, and that depends on detector width.

2:06

Many of us now are used to CT scans with multiple

2:09

detectors, with very thin, small detector sizes

2:13

in detectors, but that wasn't always the case.

2:16

And so as we approach isotropic

2:18

imaging, because we're never quite there

2:21

perfectly, it's really a

2:24

testament to the progress of CT.

2:28

To tell you how far we've gone,

2:29

this is the first CT scan done.

2:32

The first-generation scanner.

1:35

CT of the head, which alongside the fact that it

1:38

appears very grainy, took a long time to acquire.

1:42

Parameters are 3x3x13mm.

1:45

That's the in-plane spatial resolution is 3mm.

1:48

Again, it wasn't very difficult to get

1:49

3 millimeters in-plane, but through

1:52

plane, slice thickness is 13 millimeters.

1:54

Where are we today?

1:56

Here's a curved MPR of the coronary artery, which,

2:00

by the way, is something you can only do well.

2:04

That is 3D manipulation and reformation

2:06

because we have isotropic imaging.

2:09

It's 0.4 by 0.4 by 0.5.

2:09

2:10

46 00:02:10,835 --> 00:02:11,425

2:11

So the degree of improvement

2:13

is by 100, but not just that.

2:16

It's the fact that we're able to deliver this with

2:19

a reasonable radiation dose.

2:23

One of the reasons why coronary CTA took off and

2:25

coronary MR didn't is because of spatial resolution.

2:29

Coronary MRA, actually the first papers for

2:32

coronary MRA came before coronary CTA, but

2:35

even now with the 3 Tesla, which is very much

2:38

a powerful signal-to-noise magnet, you still

2:42

struggle to get less than three millimeters.

2:44

I mean, you can, but that might

2:46

increase the scan time too much.

2:52

MR, spatial resolution and scan time are related

2:55

because to increase spatial resolution, you

2:57

would need to increase the number of slices

3:01

or partitions, as I said, and increase the

3:05

phase matrix, increase the phase matrix.

3:08

You increase the scan time.

3:10

So you get a penalty.

3:12

CT, spatial resolution and acquisition and

3:15

temporalization are mostly unrelated, mostly.

3:19

So this is what happens when

3:21

you don't have isotropic voxels.

3:23

You get what's known as a stair-step artifact,

3:25

which is fine if you're looking at the femur.

3:28

It's not so good if you're looking at the

3:30

coronary artery where those stair steps

3:32

themselves can be a source of confusion and

3:35

artificially create stenosis when there isn't any.

3:40

So to recap.

3:42

Coronary arteries are tortuous.

3:44

They don't travel along a single plane.

3:46

They go in and out of plane.

3:48

So in order to be able to see an

3:51

artery, for example, at this point here and

3:55

know exactly what the degree of stenosis

3:57

is, you can't cut it axially, coronally, or sagittally.

4:02

You have to go along the long axis

4:04

of the artery and cut through it

4:06

orthogonally.

4:08

So it's a two-step thing, but you're only able to

4:09

do that if it doesn't matter which plane you're in.

4:13

So it doesn't matter if you're in

4:14

axial, coronal, or off-axial plane.

4:16

The fact that you have isotropic voxels

4:20

means that there's no informational loss.

4:23

And it's not just good having isotropic because you

4:27

can have isotropic imaging with five millimeters

4:29

by five millimeters by five millimeters.

4:32

The point is you want to have isotropic

4:33

and thin, not just isotropic, and both of

4:37

which are necessary for coronary imaging.

4:40

So how has the spatial resolution improved?

4:42

And I think that this is a lecture in its own right.

4:47

You can even write a book on it.

4:50

Two things to appreciate.

4:50

The first is that the detectors

4:52

have become much more efficient.

4:55

The power of the CT scan has increased.

4:59

So we're delivering more iodine flux, but

5:02

we're getting more out of that iodine flux.

5:04

So we're not wasting radiation.

5:06

So in a sense, things are more powerful,

5:09

but also things are more efficient.

5:10

So power and efficiency.

5:12

Has resulted, along with multiple other things,

5:16

in us being able to do isotropic imaging without

5:20

hugely penalizing in terms of radiation dose.

5:24

Thank you.