0:00

Okay, this next case is an example of one

0:03

of the pitfalls of using CPR images.

0:07

So CPR images their curved planar reformatted

0:10

images the computer software traces what

0:13

it thinks is the center line

0:16

of the vessel and connects the

0:19

dots all along the length of the vessel and then creates these

0:22

reformatted planes that basically

0:25

are looking at the vessel and for multiple different

0:28

angles with all angles sort of moving through

0:31

the center of the vessel.

0:33

So with those cbrs and you can spin around the center of

0:36

the vessel and get all these great images that we've been looking at.

0:38

A problem that can happen is if the center line

0:41

tracking is inaccurate, then we can get pseudos to

0:44

noses.

0:45

So take a look at the RCA here.

0:48

Looks perfectly normal on this.

0:51

Angle, but then as we scroll around

0:54

you can see this.

0:57

It looks like a stenosis and if all you

1:00

had resist picture.

1:02

You would think for sure that there was a moderate stenosis in

1:05

the mid RCA look kind of fuzzy. Maybe it's from non-calify

1:08

Black.

1:09

And you know, I couldn't fault you. I say the

1:12

exact same thing.

1:14

But what should make you question is how it's so easy

1:17

to scroll in and out of it and then their areas of the perfectly normal.

1:20

And so this should raise your concern

1:23

that maybe this is a tracking problem.

1:27

And so then you go back to the raw data.

1:30

in this case we can pull over the

1:34

stack of thin images

1:36

and look at that region of the RCA, which is right here.

1:40

And I don't see anything looks perfectly

1:43

fine.

1:44

Right, and then you can also do the short axis

1:47

reconstructions.

1:49

and on these short axis reconstructions

1:52

We can also take a look at that same area.

1:56

And if we follow the right coronary from proximal.

1:59

to distal

2:01

nothing

2:03

And so what's happened is that there's a little

2:06

bit of blurring for motion proximally, but that's not the same area down

2:09

distally here. There's nothing so what happened is

2:12

again, the curve cleaner images just

2:15

sort of hopped out of the center line a

2:18

little bit and then hop back in and it can give you this pseudo

2:21

stenosis. So don't rely completely on your

2:24

CPR is always make sure to the cprs and

2:27

your Source data match.

2:29

And you know that'll prevent you from

2:32

making any mistakes. Now one of the thing which I

2:35

think you can see a little bit in the LED here that

2:38

can also happen with

2:40

curve planner images is that oftentimes it'll

2:43

have trouble with tracking when you

2:46

have a really tortuous vessel. So if you look at the

2:49

Led here see there's that little potential stenosis right

2:52

there.

2:53

What that is is it's just a curvy segment of the LED

2:56

when the LED curves a lot like this tracking as

2:59

a density to sort of cut the corner and go basically around

3:02

the edge and when it

3:05

does that if you get the correct angulation, it'll make

3:08

it look like there's a stenosis just

3:11

like that and it's not a true stenosis again. It's just a torch with

3:15

segment and this one honestly is even that tortuous but

3:18

a curved segment can sometimes result in this other

3:21

type of CPR artifact that we see here. So those

3:24

are things you need to avoid to make sure you do not accidentally

3:27

cause something positive for lesion one. You have poor tracking

3:30

the CPR.

3:32

I just notice actually one as we're going through it. Looks like the middle

3:35

idea has some tracking problems see

3:38

that.

3:39

So I don't know what was going on with the computer this day,

3:42

but it was not doing good job with the CPR and we

3:45

have another area of students in the middle ID

3:48

in this particular patient, which if we go back to the raw

3:51

data.

3:52

We can see that the mid LED is is

3:55

perfectly fine.

3:57

Especially when taking a look at that region with the myths.

4:02

So that's an example of pseudosynosis related to

4:05

poor Centerline tracking on the automated CPR

4:08

images.