0:01

Okay.

0:01

We're now gonna discuss, uh, tomosynthesis

0:03

approach to interpretation and lesion localization.

0:08

The approach to interpretation for, uh, DBT, uh,

0:12

is in large part very similar to the approach

0:14

to interpretation for standard 2D mammography.

0:17

Of course, there are some differences

0:18

incorporating the DBT dataset into your

0:22

search pattern or hanging protocol.

0:24

Um, but all the other stuff is the same, right?

0:26

We need appropriate environment for

0:28

interpretation that includes low-level

0:30

lighting, no interruptions, hopefully.

0:33

Uh, and in some practices this can be done with

0:35

batch reading as well. As with 2D digital mammography,

0:40

you need a standardized

0:41

approach to interpretation, so

0:43

everybody can come up with their own, um, sort

0:45

of way that they approach an imaging exam.

0:48

This is true for anything we do in radiology, um,

0:51

but it's important to do the same thing each time

0:53

you do it, uh, and maintain your same search pattern

0:56

as you're going through screening exams.

0:59

As you all know, screening mammography in

1:01

general is a relatively brief exam, right?

1:03

We spend maybe one or two minutes

1:05

looking at a screening exam, and so

1:07

we need to be able to do that quickly,

1:09

uh, efficiently, uh, but make sure we also

1:11

view all of the different portions of the exam.

1:14

So, um, for me, what I like to do

1:17

is look at the sort of current exam.

1:19

First, I look at — um, we use, uh, SM in our practice.

1:24

So I look at SM views, sort of a global

1:26

picture about what the exam, uh, looks like.

1:29

I like to look for things like image quality first,

1:31

make sure the patient is well positioned, make

1:33

sure the images look like they're pretty good.

1:35

Um, and then I like to compare to the DBT views.

1:38

I go through the DBT slices for

1:40

each of the four different views.

1:42

I might review additional views that have been obtained

1:45

by the technologist to get more tissue on the exam.

1:47

For example, an XCCL or tail view, or something

1:51

about the IMF, or maybe a repeated one for motion.

1:55

And then I like to compare it to prior exams.

1:57

Um, and I do that in two ways.

1:59

One way I do that is first to set up

2:02

the SM or DBT view on, uh, one monitor.

2:06

And on my separate monitor, I'll have

2:08

prior views that are automatically loaded

2:10

for the past, let's say three to five years.

2:13

Um, then I'll be able to scroll through those quickly.

2:15

But then I also like to do it where I'll have the

2:18

SM view on the far left-hand side of one screen, and

2:21

then multiple prior views extending to the right.

2:24

And those are static,

2:24

so I can look across the monitors

2:26

to see, uh, multiple prior exams.

2:29

Of course, like I said, everybody will

2:30

have their own way of interpreting.

2:32

Um, and that's just the way that I do

2:34

it, but I do it the same way each time.

2:36

So as I mentioned, um, I would start with an overall

2:40

view of synthesized mammography, um, and getting

2:45

sort of an overall picture of what's going on.

2:47

Right?

2:48

Is positioning okay?

2:50

Is the image quality

2:50

okay?

2:51

Are there any other things that I need to know

2:52

about before I start looking at the DBT portions?

2:56

This is the, um, comparison that I mentioned

2:59

that was sort of a static comparison, right?

3:00

So I have current DBT in the far left-hand

3:02

side, prior SM view, the next prior SM view,

3:06

and the next prior SM view from that, right?

3:08

So we can look back, assuming we have

3:10

this patient was screened every year.

3:12

We can look back three years very quickly,

3:14

same view, see if there's anything changing —

3:16

the tissue pattern, the positioning, or if

3:19

there's anything new that has, um, popped up.