How AI accelerates workflows and patient cancer care

Hi and welcome to our video cast

today we are talking about mammography

and how AI can help accelerate productivity

and patient care on your moderator, Rhonda

Blaschke and with me I have two great

resources from both Change Healthcare

and I CAD.

And one of the things we want to talk about today is how

the healthcare industry continues to learn more

about the value of learning and AI

and how we empower those radiologists.

We want to talk about how artificial intelligence

in particular and mammography is

helping speed up breast cancer detection,

improving the workflows of the radiologist,

analyzing pattern pattern recognition

and obviously most importantly helping us really

all work towards providing better patient

care. Um so today

I have on our team, Archie Mariani

the chief product officer of enterprise imaging

for Change for Change healthcare. And

in her role Archie provides product

leadership for global teams building

the next generation cloud, native clinical and workflow

solutions and she is passionate about

healthcare And has devoted more than

20 years to the field.

Archie, do you want to say hi

thank you, Rhonda, I am so happy to be here

with you and Rodney

thank you And Rodney, we

just mentioned you but you're definitely one of our

key product presenters here today. So

Rodney Hawkins, the vice president of AI

solutions that I can incorporated and

for the past six years Rodney has led Ike

has a I product strategy creating

a market leading portfolio breast

AI solutions

that are being used worldwide to benefit both

providers and women in the global fight

against breast cancer.

He spent over 25 years focused

on developing and introducing innovative

technology to improve clinical

care and efficiency and diagnostic

imaging and we couldn't be happier

to have Rodney and the team and I can

as partners with us Rodney do you want to say

hello, great to be here. Hello everyone and

thanks for inviting me,

artie. I'm gonna start with you first on our discussion

and I'm gonna ask you a little bit about what

your personal background is and your expertise

and and what drives you. Why is

mammography so important for radiology

customers and how we're actually

helping to drive the digital transformation

and healthcare from an enterprise imaging perspective.

Can you tell us a little bit about that?

Thank you Rhonda. Yes, for sure. I mean I think

a lot of people don't know but I um

I actually have a personal story

of you know why even you know change

and and how it actually,

you know, it's connected to my mom

and her experience with breast cancer

Um personally, you know, I have

like you said, you know devoted 20/20

years to healthcare and and been

in health tech for most of my life.

And you know the reason is that I've always

had this idea of you know finding

passion with purpose and healthcare

always allows you to do that.

My mom got diagnosed with breast cancer

and we really wanted a second opinion.

She was in India and I wanted her

to get a DBT done

um which she did and I really wanted

to get a get sort of you know send

the image across so that you know, we could get a second

opinion here in the United States and

I couldn't believe that

back in 2019, there

wasn't a way for her to be able to do that

when we could, you know, when we talk

about all of the big data and the big

tables at Google, we didn't

have that ability to share um

to share an image. So for me it was a personal

challenge to be able to

develop an enterprise imaging platform

so that you know, you can truly democratize,

you know patient data and

expedite care in the best

way possible. I think any

journal that you pick up today,

you're going to find, you know true

adoption of Ai really

has only been in the space of tomography

and you know partners

like Ichat innovators like ICHAt

are truly making that possible when you talk

about radiologists

For them having to do 17,000

different things for one study,

one read

To be able to elevate that

productivity and reduce the burnout

by 70%. Those

are sort of you know, even conservative numbers that

we look at, you know of any study out there. It

doesn't matter whether it's an Ichat

um you know algorithm or something else.

So to be able to truly

focus a on our patients,

you know, to expedite care when they really

needed and be on our providers

who are continuously burned out

to be able to make them read, you

know faster and more effectively,

really focusing on those high,

you know, highly complex cases.

I think that I genuinely feel

like AI has really arrived in the right

place at the right time.

So Archie that's such an important point.

And I think one of the things that we really need

to understand is how the workflow really

is a key driver and the ability

to provide um great

patient care. And so Romney, looking for your perspective

on what are some essential elements

that providers should look for when they're adopting

AI to streamline that clinical

workflow and really improve clinical

outcomes. Can you talk to us a little bit about what you've

seen and what your perspective is?

Sure. And that's a great question because

not all AI is created equal.

Um

you know, first is is how well does the AI

do what it's supposed to do in other words,

how well does it perform?

And there's there's two ways to think about performance.

First is

what we call standalone performance, which is,

you know, how well the AI performs on its

own. Um and this is typically

measured by running the AI on datasets

with known outcomes known

cancers to measure sensitivity,

known normal cases to measure specificity.

The second and perhaps more importantly,

is how well the radiologist

performs when using the Ai.

Um This is typically measured with

a reader study where radiologists read cases

both with and without A. I. Uh

and the results are compared ideally

the AI should improve radiologist performance

at least one of three areas sensitivity.

Or does it help them detect the disease

in the case of mammography. Ai does it help them

detect cancers

uh specificity or does

it help them reduce false positives and

ultimately unnecessary recalls

um or efficiency. And does it

help the reader do their job faster?

It's really difficult to find any

kind of technology that will improve

performance in all of these areas.

Um But but we were pleased,

very pleased with the results of

the profound AI reader study which

showed radiologists improvement in cancer detection

rates, reduction in the number of

false positives and recall rates as well

as a significant reduction and reading time.

Um In addition to performance.

Uh There are other important things

to look for. Uh for example,

how much and how and the quality

of the data used to train and test the algorithm,

for example, was the data collected from multiple

sites representing a diverse population.

Um Has the AI been validated

with imaging image acquisition

technology from different vendors?

You know, if the AI only works with one

vendor's technology for example,

this can be problematic for sites with systems

from multiple vendors,

the speed of the of

the processing a case. Um If

it takes too long to process a case,

the Ai may not be available when the study

is being read, which is clearly a problem.

Uh And and finally and perhaps one of

the most important is how well the AI

integrates into radiologist

workflow.

We learned a lot over the years about the differences

and how well AI is supported

by the various packs and workstation technologies.

There isn't a one size fits all approach

and flexibility is really important.

That's why we've worked very closely with

the Change Healthcare team to ensure

an optimal workflow when using our

AI in their packs environment.

You can have the best performing AI available.

But if it doesn't easily fit

into how radiologists

do their work it likely won't be used.

That's a great point. And I think one of the biggest things

that I took away from that is

not only is it multifaceted

and what really makes effective Ai

but it is going back to how does

it really drive efficiency

patient care

being um diverse and

making sure that it has an inclusive and strong

background of the data that

selectively used to make sure that it

is accurate. Um And I think

those are great points and you know one of

the things I want to talk about a little bit and Ron young

I'll go to you first is

when you think about the opportunities and

challenges in the mammography and imaging

space.

We know that ai is extremely important

because it is more around helping us make

sure we improve in preventative

care as well as the caregiving for those that are

diagnosed.

But when the National Cancer Institute predicted

that there could be almost 10,000 excess

deaths from breast cancer and colorectal

cancers over the next decade

as a as a direct result

of delayed screening due to the pandemic.

Um But now that the mammography screens

have started to resume how

my clinicians efficiently manage

the backlog of those who need

to be screened. Um

And what is your your your thoughts on how

that can be managed and helped by technology?

Again? Great question. Um You

know our ai our Ai technology

is particularly relevant as

it relates to recovering from the impact of

covid and breast cancer screening. Um

Our product doesn't just indicate

where suspicious lesions are in the breast

but it also assigns an algorithm

derived competence scores

which indicates, you know which cases

are more likely to be malignant. Additionally,

we just recently released

our profound

Ai risk product which is a short term

risk model uh to help

identify which women who

are the women who are most likely to

be diagnosed with cancer before

their next mammogram.

You know both the Ai scores and the risk

scores can be very valuable in

helping providers identify the

women who will benefit most by not delaying

their screening appointments or those

who should be at the front of the line when it comes

to scheduling their next screening appointments,

that's really awesome. Um and as

a female, I'm very happy to hear that as well.

Um because it's definitely something that I think is of

concern with all of us. Um

so I think that's that's great to hear

before we go and I know we've asked a lot

of questions first of all, Archie, do you have any

last thoughts before we close

out this session?

I think, you know, to to what

Rodney said, it is so critical

to understand how, you know, covid

19 has really

upended, you know, radiology

and especially mammography as we understand

and as we knew it right with,

you know, sort of reduction in elective surgeries,

continuous pressures on our

radiologist to read more with

less time I guess, and fewer tools,

you know, as they as they transitioned to

working remote. There has been

extraordinary pressure

on our providers to actually

continuously provide,

you know, the diagnosis at the right time,

through the right technologies available.

And I think that's where, you know, we have to

really look at, you know, sort of these confluence

of forces that are creating, you

know, really until unreal

pressures on our radiologists and

to be able to do that, we have

to start leveraging, you know, cloud

nativity that can allow this

remote access and the

ai algorithms combined with it

should be, you know, such a great

way of, you know, talking reducing that

burnout and so with those

two forces combined, you know, or technological

advancements combined, I do

believe that you know as the elective

surgeries or government coming up, you

know the the number of radiologists

joining the workforce or in the workforce

are not increasing. And to be able

to you know sort of do the same

or more with less, we have

to start looking at how do we create

you know, these

bandwidths

of you know, productivity

for our radiologists and cloud native

enterprise imaging platform which is what

you know sort of uh for us

you know that is the changes future and

and our providers future. We're so

excited to have partnership with Rodney

and his team to be able to really allow

um sort of work for our radiologist

to work in this new normal if you will.

That's awesome.

So before we go I want to say thanks to Archie

and Romney for being part of the program.

Um I hope that all of you watching

have enjoyed the information we shared

and learned from it on how we're looking

to drive towards adoption of ai in mammography

and how we're partnering together to really drive

improved patient care. If you have

a chance to stop by in in Chicago

at RS and a change healthcare will be there

to answer any questions as well as the

eye can team. We both happen to be

located in the South Hall. We also

have a virtual booth available if you're unable

to make it in person. But for more information

also from my cat, you can visit them at icann

med dot com or in person

at the South Hall meeting suites room to

14 to 2 16.

Thank you again, Rodney and Archie really

appreciate it. Um and we look

forward to hearing from any of you on

questions you might have on how we can partner

and help you with your mammography solutions.