AI in Radiology: Are Radiologic Technologists Being Replaced or Redefined?

If you’ve ever felt your stomach drop when someone mentions AI in your department, this episode is for you.

Chaun sits down with Jordan Hermiller, MHA, RT(R)(QM), CPHQ — a radiologic technologist and Technical Manager at Agfa Radiology Solutions — to cut through the noise around AI in radiography. Jordan has been a working rad tech, a radiology manager, a university instructor, and a published researcher. Now he works at the intersection of clinical practice and real-world AI implementation. He gets both sides.

In this episode:

• Why AI feels so dominant in radiology right now — and what’s actually driving it

• The difference between real AI and rebranded automation — and how to tell them apart

• What “autonomous x-ray” actually means in a real clinical environment

• How workforce shortages and AI intersect — and what that means for your job

• New roles opening up for radiologic technologists as AI becomes standard

• When to trust an AI recommendation — and when to push back

This episode is brought to you in paid partnership with Agfa Radiology Solutions.

Connect with Jordan Hermiller: LinkedIn

Learn more about Agfa Radiology Solutions: Website Link

AI (CE) Webinar: https://agfaradiologysolutions.com/campaigns/https-agfaradiologysolutions-com-ai_impact_in_radiography/

RadX Case Competition: https://agfaradiologysolutions.com/campaigns/radx-radiology-administration-challenge/

Agfa Website: https://agfaradiologysolutions.com/

Links referenced in this episode:

1. ahra.org
2. asrt.org
3. clevelandclinic.org

This episode was created in paid partnership with Agfa Radiology Solutions. All opinions expressed by Chaundria Singleton are her own.

Keywords: AI in radiology, radiologic technologists and AI, autonomous X-ray technology, AI job impact in radiology, future of radiology jobs, AI implementation in imaging, radiology workforce shortage, AI and patient care, technology in radiology, Agfa Radiology Solutions, AI education for technologists, radiology imaging advancements, deep learning in healthcare, AI misconceptions in radiology, quality assurance in radiography, challenges of AI in radiology, AI and patient outcomes, continuing education for radiology techs, radiology automation, AI training data and accuracy.

Let me ask you something and I want you to really sit with it for a second.

If your department told you tomorrow that they were rolling out AI, would you feel excited or would your stomach drop?

Because if you felt that drop, you're not alone.

I have had radiologic technologists DM me, pull me aside at conferences asking things like, sean, is AI coming for our jobs?

And here's the truth.

I didn't have a great answer.

I didn't have one that I really felt confident in.

Not one that was coming from someone on the inside of how this technology acts actually gets deployed in real imaging departments.

I went and found someone now, full transparency.

Because that is how we do things here on A Couple of RAD Techs.

This episode is brought to you in partnership with Agfa Radiology Solutions.

They sponsored this conversation.

But if you've been listening to this show for any length of time, you know, I don't let anybody buy my opinion.

What I do let them do is bring people to the table worth talking to.

And Jordan Hermiller, worth talking to.

So let's get into it.

My guest today started exactly where you are in the room, running fluoro, positioning patients.

Then he became a manager, university instructor, published researcher.

I know the list is going on and on.

He is worth the wait, you guys.

Then he went to the other side of the table and now he's managing the North American team at Agfa Radiology Solutions, a company that uses AI imaging, where he works directly with the technology, the clinicians using it, and the teams being trained on it.

He's been a technologist, a manager, a teacher, and now sits at the intersection of clinical practice and real world AI implementation.

He didn't build the machine.

He knows how to use it, evaluate it, teach other technologists like you and I how to think about it.

Which means when he talks about AI in your department, he knows what it's like to be you.

Jordan Hermiller, RTR QM so many certifications.

Welcome to A Couple of RAD Techs, Jordan.

Thank you, Chandria.

And that was an unbelievable introduction.

I've never had an introduction like that before, so I really appreciate it.

Happy to be here.

You are welcome and you are deserving.

We met at RSNA and I got to see Jordan in action.

I'm talking from experience.

I was blown away at rsna.

It was my first time going, thank Akfa for inviting me.

But I also got to see a company that uses technologists.

I mean, I don't think I've ever seen that before in the positions that you guys are all in, most people have been where we are.

They can speak from a technologist role.

Jordan, give me the real version of how you got here.

You know, I know you're doing some great things for an amazing company and for our profession.

Tell us, how did things go from radiologic technologists to where you are now?

Sure.

I love being asked this question.

What got me interested in, in radiography was they have a couple of phenomenal aunts who are nurses and they actually were talking about healthcare and I had an interest in healthcare and helping people and they suggested radiology.

And I think like a lot of us who are technologists, there is that kind of combination of we're interested in healthcare, we want to help people, but we're also interested in technology.

And it's really kind of a perfect marriage, if you will.

And so that's what got me interested in in radiography and radiology.

I went to Ohio State and I graduated with a bachelor's in Rad sci.

And from there I worked as a technologist for several years at Ohio State at an orthopedic clinic.

And then I had an interest in leadership.

I went back to school to get my master's in healthcare administration and at that time actually started to teach as well, which I still do part time today.

I thought I was going to work in the hospital the rest of my life.

That was the goal of going back to school even, you know.

But I knew someone who knew someone who knew someone was leaving at agfa.

And so they say, would you be maybe interested in this type of role?

And at first I thought, no, again, it can be so easy to, I'll say pigeonhole, but put yourself.

And this is what I want to do, you know, I know it's a lot, a lot of what you talk about as well.

I started really thinking about the job, thinking about the job and it sounded like really a great blend of my interests of, you know, one leadership.

But even like being able to be creative with where radiology can go, the future of radiology and X ray, how you can actually help influence technology to help technologists to help patients.

And I thought this actually sounds pretty awesome.

And so I've been with AG before, coming on five years now and I tell you, I don't regret it at all.

I really love it.

We have a great team and you really get to use a lot of different skills, skills working on, I suppose, the vendor side as well.

That is a wonderful opener for why you are so qualified to talk about this.

Conversation.

You, you come from a technologist world, but I love how you pivoted in your career.

But I do want to talk about that AI.

AI is not new.

It's kind of been in the conversations everywhere.

The radio, especially radiology conversations.

Computer Aided Detection has been around for years.

Why does it feel like AI is everywhere in the radiology conversations right now?

What's changed?

Yeah, I mean, great question, Chandra.

As you mentioned, you were at RSNA four months ago already.

Hard to believe, feels like it was yesterday.

But you can't walk more than 10ft at RSNA without seeing AI plastered everywhere.

And it's not just RSNA.

Radiology Publications has voted AI the most significant news event in radiology for the fourth year in a row.

As you mentioned today, AI in radiology is still primarily focused, you know, on radiologists, helping with their interpretations, helping with pathology detection.

My work at agfa, my focus, and our focus is at point of care.

It's helping technologists be more consistent, more efficient.

Does AI exist at point of care today?

Yes, it absolutely does.

I also think we're kind of really on the edge of what's possible and where we're going, which I know we're going to talk about.

You know, maybe more specifically, to answer your question about why now?

Why is this seem to be blowing up now?

I think there's really two main factors that are converging.

One is this workforce shortage.

And the second is the technology itself, the AI technology itself, which is expanding.

Regarding the workforce shortage, if you were to survey any radiology leader and ask them what their primary challenges are, they're going to say financial pressures and the workforce shortage.

ASRT data, they said that vacancy rates tripled in just a few years from 6 to 18%.

Again, I think that's one of the reasons why what you're doing is so getting people excited and more excited about being a radiographer, being a technologist.

But AI can impact essentially anything, any type of workflow, and so it can help meet this workforce shortage.

That's part of the answer of why it's exploding now.

The other part, again, is the technology itself.

I'll try to keep this somewhat high level, but I do think it's important, you know, to start to understand what is AI, how it can help, and again, why it's kind of taken off so recently.

There's kind of three things.

One is the advancements in what's called deep learning.

Okay, deep learning is actually a, a type of machine learning.

And so maybe just to explain this a bit more Machine learning explains how AI learns, which is from data rather than being explicitly programmed with traditional machine learning, a human still has to tell the AI what to look for and what to measure.

Deep learning actually skips that step entirely.

Essentially, you give it the goal and the examples and it develops its own features to look for.

Right?

These are things that us humans would never think to measure, right?

That's what makes it so powerful.

In medical imaging, for example, you know, an engineer, if they're trying to code for a pathology detection, they could code, measure cortical thickness, but no human can code, you know, detect every subtle fracture across every variation of human anatomy.

Not possible, not realistic.

And deep learning can find these things that we wouldn't even know to look for.

The one aspect is deep learning, another is this explosion of big data.

When you think about all the data we create every day from, you know, social media or smartwatches, even health, health records make up 30% of the data that's generated every day.

As we know, AI thrives on this data to become more effective, more accurate.

And then the last piece is the improvement in computing power.

There's any gamers listening, you know, these, these GPUs, right?

They were actually created initially for gaming, but they turned out to be highly efficient for training these AI models.

And so since then, companies have developed these chips specifically for that purpose.

The technology has advanced so rapidly recently because the algorithms, the data and the computing power have all come together.

And then on top of that, you also have to consider radiology was really built for AI.

You know, if you think about how technology driven our field is, one, two, there's, we already had hundreds of thousands of digitized X rays that have labeled pathologies that are perfect, perfect, you know, training data for these AI algorithms.

But what's telling is over 75% of those are in radiology.

And second, a number of them have what's called an add on payment designation from cms, which again, actually is pretty important because that tells us that these aren't just kind of fun, interesting technologies.

These are actually proven to offer real value in clinical care into patients.

Maybe just to kind of sum it up, you know, one, it's the workforce shortage being able to impact that.

Two, the technology has advanced so rapidly somewhat recently.

And lastly, radiology was really built where AI can excel in this area.

When it comes to technologists, because we're talking to technologists, what do they misunderstand most about AI and radiology.

Sure.

I think, I mean, I think that's also a great question.

And I think honestly one of the biggest misconceptions when it comes to AI in radiology is actually what is AI?

Right.

You have to consider there's a wide range of definitions that's used by industries to define something as AI.

And with all the buzz around, AI vendors, you know, can take advantage of that ambiguity in their marketing.

Right.

For example, if you define AI as anything that mimics AI, like if you were to Google, what is AI?

One of the definitions is something that mimics human intelligence.

Right.

Human decision making.

Well, if that's how you define it, then you can qualify a lot of features as AI actually.

Right.

I mean, I think one, one good example is the automatic exposure control, the aec.

Certainly the AEC is used to help mimic human decision making.

Now, as technologists, yes, we should always be responsible for our techniques, but of course it'd be unrealistic to say the AEC doesn't take some of that decision making from us.

AEC has been around for decades.

Undoubtedly if it were released today, someone would slap an AI label on it, call it intelligent dose or something.

But it's not true AI.

It's not machine learning.

And actually I give a presentation on AI to radiography programs to hospital departments as a CE approved AI kind of webinar.

And what I start with in that presentation is actually kind of a fun story to illustrate this difference between true AI and kind of traditional rule based computing.

I tell a story about two robot friends that you have.

One's name is Norm for normal computing and one is Artie for, for artificial intelligence.

And both Norm and Artie, they're great at following rules, but where they're different is Artie is also a great learner.

So you can show him examples and he can use those examples to figure out different situations on his own.

And that's quite important in radiology.

When no 2x rays are the same, no 2 patients are the same, that Arty can really be quite valuable.

And then I also build on that by kind of telling a story about, you know, Norman, Arty also like to read books.

Hang with me here.

It's Where's Waldo?

If that counts as reading.

And so, you know, you can show both Norman Artie thousands of pages, Where's Waldo?

Pointing out exactly where Waldo is on each page.

The next time they look at those pages, they can immediately tell you where Waldo is.

It's so impressive.

You invite your friends over to see him in action.

But eventually there will be a new publication.

Awares Waldo.

Your friends bring over this new book.

They hand it to Norm and already, well, they hand it to Norm.

But sadly he's lost and confused.

He has no idea where Waldo is on these new pages.

Maybe it's because he was simply programmed on the coordinates of where Waldo is.

But how is that programmer going to know where Waldo is on these pages he's never seen before?

Next they hand it to Artie.

And Artie can still identify Waldo on these new pages that he's never seen before.

Well, how is that possible?

Because he was actually trained on what Waldo looks like.

The red and white sweater, the glasses, the hat.

Artificial intelligence works a lot like our friend Artie, where he can, where it can learn and adapt to new situations.

And once it's trained, you know, it retains everything it's learned.

It can can read images lightning fast and it never, unlike us, it never needs a coffee break, never gets tired.

And then maybe the other thing I'll mention in that presentation is I also about halfway through we play a game called Is it AI?

Where I share a number of AGFA features that are part of our Smart XR portfolio.

And I ask is kind of take a survey, a live survey of, you know, I explain the feature, I say do you think this is AI or not?

And I think it's quite eye opening actually because a lot of the features that you think oh wow, that's quite impressive, that's very helpful.

There's vision based cameras that show you, help you position and so on.

And they say oh that's AI but it's not.

And then sometimes some of the more what you might might consider mundane features like the feature that automatically rotates the image.

But by that point they're like oh, that's not AI.

But it is because if you think about how you would have to program something like that, you have to feed it hundreds and hundreds of images so it can learn the correct orientation and orient the image correctly.

Because no 2x rays are the same, you actually have to use artificial intelligence for some of those features like auto rotates.

I guess again, to sum up a long answer, I think that one of the biggest misconceptions is just what is AI?

I think vendors can blur those lines to their advantage to some degree.

And to be fair, there is some features where there is a true blurring of is it AI or not?

Where it obviously uses rule based automation and AI.

But yeah, I think it can matter whether it's A or, or not, maybe this is the last thing I'll say it can matter whether it's AI or not.

Once this AI is released into a radiology department, it's, it's locked, right?

It's not continuing to learn from more and more X rays.

And that's actually important for a couple of reasons.

You know, one, the FDA requires that it be locked for safety purposes, right?

You have a validated algorithm that you don't want to change over time.

The second reason it's important, I think important to technologists is operationally it can be quite difficult to discern between what is rule based automation that we're used to versus what is true AI.

Because you consider once it's released and you're using it, both normal computing rule based automation and true machine learning AI, they both accept inputs, apply their internal logic and produce a consistent output.

Right?

The real difference between rule based automation and true AI is the type of problems that they can handle and help solve.

I really love the Where's Waldo that gave a visual that I think most people can compare when it comes to act for radiology solutions.

You guys are in my opinion pioneering AI in technologist roles to clear up the confusion because I feel like that is anybody can kind of brand stuff and we do get kind of confused like is this AI or is it just automation?

How can text tell if something is in the most simplest way?

Like as a tech on the floor, how can I tell if.

Because ac, I mean I get that that makes sense.

What's something that nowadays every technologist, especially the new generation Gen Z, even the millennials, maybe they didn't go back far and kind of can remember the AC or really doesn't.

They can't tell the difference between that and something else.

They didn't come from a film background, so how all they've known is digital and sometimes people get that confused with oh, that's AI.

How can a technology, someone Gen Z's entering the profession really understand how when something is true AI so, so they're not scared, so they're not confused.

Well, I think as I was kind of trying to allude to, it's not easy because they both again the AI, once it's in the system, once you have a feature that's released, it's not continuing to learn and adapt and change whether it's rule based or AI, there, there's, they're very useful features.

Right?

Regardless if you wanted to truly know whether something was AI or not, I mean there's some questions you could ask the vendor in terms of how it was programmed, how it was trained, like was it trained using actual clinical data for example, and was it validated on actual clinical data?

Did it actually learn to kind of handle new situations on its own with this clinical data?

That could be one way.

Certainly you, you can critically think about these things, how, how it likely could even be programmed, what kind of problem it's trying to help solve.

I know not everyone has access to be able to do this, but you, you know, if you can talk to vendors and ask them these, all these pointed questions that does kind of help parse apart whether it is rule based automation or true AI, sometimes it can be quite hard to reliably know.

Now I've got a chance at RSNA to actually touch and feel agfa's equipment and see the AI and the automation.

For myself, I was blown away.

You can talk specifically how you guys AXFA clear up that confusion when it comes to your use of AI?

Yes.

I don't know that we spend a lot of time really trying to discern this is AI.

This isn't.

I think the goal for the end user and the technologist is this is a feature that can help you.

This is how it works.

Now certainly if they want to ask questions, to dig into the weeds, I love having those conversations, you know what I mean?

But you know, beyond that, certainly we are using AI here in certain ways.

You know, in image processing with the, with the noise reduction and the virtual grid and the segmentation with auto rotating images with helping positioning.

The question is how do we tell them whether they're using AI or not?

I think what's also can be important as AI continues to be rolled out though is vendors I think do have responsibility for some of these AI features where I think not so much today, but I think it's coming without overly disclosing.

There has to be a transparency between, you know, what does this AI do?

Obviously what are the strengths but also what are the limitations.

When we talk about maybe like when to trust or not trust AI, I think it can be important to understand what is AI because that can play an impact on how you critically think about some of these features when you're seen.

I love that answer.

Now we're going to get to the autonomous X ray part because we all watch sci fi and we see in the news when, when technologists hear autonomous X ray, they picture a room with no technologies, just a machine doing everything.

Is that actually what it means?

People be freaking out.

You might be asking because again you're at rsna and, you know, at rsna, we, agfa, announced our vision for autonomous X ray.

To your point, though, something I definitely want to, I definitely want to stress is.

Sounds scarier than it is.

I think when people these days, myself included, hear AI autonomous, it's like, oh, boy, what does that mean for me?

What does that mean for my future?

What does that mean for my kids?

When we're talking about the autonomous X ray, you know, we're not talking about autonomy from people, we're talking about autonomy for people.

And what I mean by that is, you know, I already touched on this workforce shortage.

And another area that people don't consider a lot of times is this decentralization of care.

Care is moving from hospitals, from cities to rural areas and to outpatient clinics.

I mean, and the reason for that is to bring care closer to patients and closer to our aging population.

There's a good reason to do that.

But the problem is when you hear 30% vacancy rates, that doesn't mean 30% vacancy rates across every part of the country.

When you're in a lot of these rural areas, that means 50% vacancy rates.

Potentially, when I say autonomy for people, what I mean is how can we maintain the quality of care that patients need deserve with this real staffing problem?

Today we are using AI, as I mentioned at point of care, to help technologists to remove some of these mundane tasks, to allow them to essentially to do more.

They don't have to rotate images.

They might not need to collimate.

They don't have to do as much window leveling because of the AI and the image processing.

But what I think is also important to understand is that is not enough, right?

When we're going to have four years, three and a half years, potentially, we're going to have 30% vacancy rates.

That is not enough to meet the patient demand and give them the quality of care they deserve and to ensure that they're not losing access to radiographic exams.

That's a real problem.

And these AI features here and there alone are not enough.

But what we're talking about, the autonomous X ray, again, at least agfa's vision, is this would be a room that handles a very specific scope, a system that would really focus on just chest X rays, like PA and lateral chest X rays on patients that are proven to be ambulatory, communicative.

It would be able to handle those types of patients with minimal technologist oversight.

Alrighty.

And even there, just to draw a comparison, you might think of like a the supervised self checkout at the Walmart, which I know people hate.

I understand that.

But the idea is to again extend the capacity of technologists so we can meet the patient demand.

0% are is the is they're looking to like, hey, we can do the job of a technologist?

Absolutely not.

It's just we, how can we realistically help so we can continue to provide the quality of care that there needs to be?

You know, how would this, this technology work?

Will there be safety checkpoints along the way?

Of course, this, the system would have to confirm, oh, this is the correct patient.

I've confirmed that they can, they have the mental and physical capacity to kind of position themselves for a chest X ray.

The room would be able to automatically position, which of course some of these things obviously we already have today.

Right.

It would need to automatically collimate, you know, look at the thickness of the patient, adjust the technique.

Of course we have AEC and something that is also here and coming is potentially doing automatic quality assurance.

Right.

But all the while it might mean the tech can manage two rooms at once or have a, have a patient their, their imaging and then this other system would say, would have give a some type of notification to say, hey, the patient is ready, you go in, you check the images, do you have any questions?

And everything looks good, they can go on about their way.

But you know, beyond that, I think what's important, really important to understand is, you know, of all the imaging modalities, X Ray, MRI, CT and so on.

One of the hardest things to automate is always going to be X ray positioning, which we know is one, not easy and two, critical for a diagnostic study.

And so there's just no feasible future, near future where AI is going to be able to position, you know, a patient in the edge for a axillary shoulder who can't abduct their arm.

There's no realistic future where it's going to take a portable and into the ICU and actually a non response, you know, non responsive patient.

I mean, so, you know, we're talking about a very, again, a very specific scope of patient to try to realistically meet this future demand.

And the technologist remains absolutely critical, you know, within that vision.

And really it even allows them to practice at the top of their license.

Again, you know, even in the example I gave, they don't need to be clicking the rotate Button, you know, it allows them to focus on areas where only humans excel, which is being present with patients, showing them empathy, making sure their questions are answered, and doing some of these more, you know, the more difficult positions and type of exams.

There are so many other things that we could be doing, you know, that would make our job and the quality of care much better.

And I feel like you really gave us, you know, you align our fears, you relax those, because we got to really see what autonomous X ray really looks like and what AXFA is doing, ACTA Radiology Solutions is doing to make sure we, as technologists, we stay central to our patient care by using AI.

I really love that.

Thank you so much for clearing that up and allaying our fears and confusion that is out there because there's just so much information and things are coming fast.

It goes back to agfa's keeping humans central.

You made a very good point in making sure that the AI that you use, the autonomous X ray is not there to remove humans.

There are so many things that we are needed there for.

So to take away some of those mundane tasks that we don't need to be doing that could be focused somewhere else.

So let's talk about patient care.

Since you touched on it, we're going to talk about it because every vendor deck, you know, I've seen talks about the workflow, efficiency, the throughput and those.

And those things matter.

But how does AI and autonomy actually improve?

What happens to the patient in the room?

You talked a little bit about it, but the scare patient, the pediatric patient, the patient who's never been in an imaging suite before, this is keeping humans central.

I think I already touched on a little bit.

Is AI collimation.

There's been published studies that show that it can reduce acquisition time from 45 to 31 seconds per exposure, which is certainly meaningful.

It also can cut the amount of times that a tech interacts with the collimation in half.

So it's again saving them time to be able to focus on the patient.

And things that matter can save 20 hours over a year on a single modality.

And that study was actually published when we only had it for distal extremities.

Today we have it for every single exposure.

Right.

That number would be even more so.

Again, that's just another example of technologist goes to do a portable X ray exam.

They don't have to worry about the orientation of the detector anymore.

They can just put it behind them and it just gives them a few more seconds at least to focus on the Patient.

I think that's a big part of Aqua's vision, is it's trying to let them focus on again where humans can excel in what matters and not on the mundane things.

Another point I'd make, too, is burnout's a real thing.

I mean, you know, you talk to techs every day.

I talk to a lot of texts.

I mean, burnout is a.

A real problem.

And if there's a lot of X rays to take, essentially now, maybe to give a couple of examples of AI that is here today, but maybe more on the horizon.

Say you had to do a portable chest X ray in the emergency department.

It's a challenging patient.

You know, it's.

There's lots of lines and tubes.

You put the sponge behind them.

You take the X ray, you look at it, and you're like, boy, you know, it's not great.

It could be better.

But I don't know if I have the time or ability by myself maybe to make it better.

So you slap the best image possible on there.

You send it on the way.

Well, turns out, you know, the.

The X ray was rotated, and so the heart is enlarged.

The asymmetry the lungs hides a potential pneumothorax.

There's been studies, and this again is certainly emerging where you can have AI.

It's basically quality assurance at the point of care immediately after exposure.

That could tell you, oh, their chin is in the lungs, or they didn't have a full enough inspiration, or maybe they were rotated.

There's an AI alert that says, oh, they actually were rotated beyond a reasonable threshold for a diagnosis.

What's considered a reasonable threshold for a diagnostic study?

You take that into consideration.

You look at the image, adjust your sponge again, you take another X ray, and it goes to the radiologist.

Oh, there's a pneumothorax.

You know, say you're doing a pelvis X ray.

Trauma.

A trauma pelvis X ray.

We do them all the time in the emergency department.

And on the pelvis, it spots a hip fracture.

You then think, oh, well, you were considering doing the frog leg lateral, but now you say, oh, displace this fracture or make it any worse.

So now you know to do the cross table shot.

So you.

You don't displace this fracture for that patient.

Or maybe you do, you know, adva actually, we release where you can use the X ray room to do tomosynthesis on with them staying in their stretcher, which can show you the information you need from the frog leg lateral without them moving at All So you can also be used some of this AI to impact your workflow for patient care and patient outcomes.

In each of those cases, the AI informed the technologist decision, but they made the decision.

Right.

And I also think, you know, I think vendors have really a duty almost when there is a technology that's going to help patients to focus on those features.

Right.

To help techs, but also to help patients.

Yeah, I was at RSNA and I saw exactly what, what Jordan just talked about.

I was on the tomosynthesis table, having a good time, but I got to see how you guys put that together.

Seamless.

And as a technologist who's worked in those kind of facilities where you just got one shot to get that board behind them and get it right.

And I've been guilty of missing and flipping images the wrong way.

I mean, who hasn't?

I hear you.

And get with vendors and ask questions when they come in.

You know, reach out to those that can answer those and lay those fears for you.

But this really helps us to see how keeping humans centered and not removing jobs is so important.

So that parlays over into an upswing in our conversation.

We understand AI, autonomous X ray, you, you, but the future roles and opportunities that come from AI.

Let's talk about what actually is opening up, because I don't want this conversation to only be about what's going away or the fears or the anxiety.

What new roles are you seeing radiologic technologists move into as AI become standard?

And are these roles that working technologies can actually access, or do they require you starting over?

Because most of us, you know, we're working full time.

We're.

We have lives.

We don't want to have to stop and start over.

But even for gen zers coming into the field, what should they do to prepare?

Maybe like, what kind of things should they be doing?

I really appreciate you asking this question.

In terms of what options will exist now and going into the future, the role is evolving.

What new opportunities will this create?

Right.

ASRT consensus committee, for example, they actually recommended that AI play a real role in radiography and that they actually recommended that these radiography programs require AI in their curriculums.

And then internationally, the ISRRT stated that techs should play an active role in developing and validating some of these AI features.

What that tells us is, you know, these radiography societies around the world, US technologists need to take ownership of these changes.

We need to be a driver of, not a passenger.

So what does that mean?

AI learns its own logic from Data, something goes wrong.

You can't necessarily just open a manual and trace the decision tree like you can with rule based automation and say, oh, this is why it's not working, this is what's wrong.

This is actually what's known as the black box problem when it comes to artificial intelligence.

Where an input comes in, there's an output, which is the recommendation for the AI algorithm, them, but you don't necessarily know what's in the sauce, what's in that black box, what's making that recommendation.

So if something goes wrong, how do you trace that?

So it can be diff.

That can be quite challenging.

To give an example, there are a lot of the X ray systems have these, these 3D cameras do like a reconstruction, a visual reconstruction of the patient.

It can be used for position positioning help to help positioning.

It can be used for dose optimization.

So you know, maybe if you take the dose optimization angle, these cameras can't see the actual patient.

It's looking at the outside, right?

So if they're wearing like a gown or baggy clothes, okay, well maybe it gives you a higher technique than it should.

That is where us technologists need to be aware of what AI is doing, how it works, how it works.

And so you can say, oh well, this recommendation doesn't align with my education, it doesn't align with my clinical, clinical experience.

So I need to be critically thinking about that recommendation.

You know, ultimately, is AI going to help?

Yes, it is.

But any technology we need to be thinking critically about, is it giving a correct recommendation?

However, I really would argue, and even argued at the time, I actually had a small publication in the ASRT magazine about this, this, that the qm, there's still a need for quality improvement, right in radiography departments.

So should it have shifted from QM to more of a QI type of certification where technologists are trained to identify quality issues and learn a systematic process for how to address those quality issues in their departments.

And there's so many examples.

You know, one of the classes I teach is on quality improvement.

And so, you know, there's just so many examples where I think other disciplines have these, some of them have these types of like quality focused certificates and programs that I feel like we are kind of missing sort of in radiography.

That's one of the reasons AGFO just this year we launched RAD X which is a case competition for radiography programs.

It's in partnership with AHRA as well as the Cleveland Clinic, who actually wrote the case.

And it's to get radiographers and these students excited about quality improvement in leadership.

Right.

And give this, give them this opportunity to collaborate and compete against other programs.

There is an opportunity for QI and a passion for QI and radiology.

And when it comes to AI, that certainly aligns, you know, when it comes to technologists who are techs today working in hospitals, you know, they can learn to build QA protocols to help validate some of these AI features.

Certainly these AI features go through a lot of testing, you know, from vendors.

Some of them require certainly FDA approval.

So they go through some rigorous testing beforehand.

Companies like agfa, we're hiring application specialists.

Right.

Maybe look at my, my journey going from a tech to a manager to working for a vendor.

I think to, to really influence some of these features and AI, certainly it's much easier and much more meaningful when you have that technologist experience and you have that background.

So that obviously is important as well.

You hit on a lot of great points.

And I feel like as technologists, understanding what we do, being an X ray tech or a radiologic technologist has catapulted you to be able to write articles about how important quality management is.

And I totally agree with you on that.

I remember when I was in school, one of my instructors was huge into quality management.

She came from a mammo background, so she really, really, like drilled it into us how important your job is as a quality manager, even though you don't have the title.

She really wanted us all to get the certifications.

But I love how you said the way we open up future opportunities to get to these places as a QI or into quality assessment, quality management or anything with AI is what you're doing now as a technologist, we work in the field, we see things that nobody else sees.

We're that person that says, okay, well, if they have on genes this or if they.

But only a human sees that, and only a human that has the experience that knows that, oh, I had to tweak this a little bit on the KB versus the MA or I had to position them this way.

You only know that because you've been in the field.

And I really encourage technologists not to think they're just pressing a button or I'm just an X ray tech.

Those skills and those little tweaks that you make that nobody knows, you can take it to all the nice fancy mechanics and they can't seem to fix your problem.

But you go to the little older mechanic and he Gets the wrench and he hits it two times and now it's fixed.

He only knows that because he's been in a field working with this issue and he knows what it takes to fix it.

And I think that's what I get from what you just said as technologists, really taking seriously the things that we do in the field that only we can do, that you can turn into roles that only we should be getting.

And I don't think people see that future, but I see it.

I definitely see it because there was in YouTube there was no such thing as AI or informatics and look at how things are going.

And I just, I was blown away by what AG for Radiology Solutions showed me.

And I wish more people in the field would do more conferences, talk to the vendors that come to hospitals, really learn about the machines that you're working on and what they're capable of doing instead of, you know, just listening to the chatter that is not informed about what we do in radiology.

I do want you to talk a little bit more on the RAD X program because I think that opens up for students in radiology technology early on, how they can start to prepare and structure their strategy career wise instead of waiting until they're a technology.

Couldn't agree more with everything you just said.

And I do appreciate the, the mechanic example.

My dad's a mechanic.

So what I also think so important about RAD X is this is a case that was created by a Cleveland Clinic radiology director, you know, who's passionate about, you know, students learning.

He's passionate about radiology.

His name is Brian King.

Great guy, really sharp as well.

So these, this is a, these are real problems that he faced in the hospital, at a very, very prestigious hospital, being in Cleveland Clinic.

And so these, these are examples that students are going to engage with, interact with that are real world problems.

It's not necessarily just a theoretical exercise.

These are real problems that radiology leaders are facing today.

Employee morale.

It's really about working with a team, you know, working with your classmates, being creative about what's going on here, what are some, you know, creative solutions and sharing those ideas.

That's the idea behind the case study.

We have a radiography program.

Who's interested?

The deadline is March 13th.

If it's past that time, you can reach out to me, we can get your team signed up.

There's also a great package if you win.

I think it's $2,000 cash prize for the team who wins in all expense paid trip to Ahra.

Their annual meeting where you'll be recognized regardless of whether you win.

Again, it's great experience.

It's going to look phenomenal on the resume, especially as this competition grows year after year after year and gets bigger and bigger.

Having that on your resume, whether you win or not is going to show you take initiative.

And it's virtual too.

You get with your team.

The deliverable is, you know, kind of like a short executive summary and just, you know, your presentation and then you give a 10 to 15 minute presentation to a group of, you know, there probably be someone from acva, someone from hra, someone from Cleveland Clinic.

You give your presentation, they'll ask you some questions and everyone receives feedback I think within like a month gets feedback on the performance, how they did, how they could improve.

So really I think a unique and a great opportunity for students that is

great to help people to see that not only AI is assisting how Agfa Radiology Solutions is using AI in radiology.

I think you said 75% of the data like our films and things is used by AI to just so much great information here.

I appreciate it, Jordan and ACT for Radiology Solutions.

But I want to kind of summarize everything that we've talked about because you gave some great analogies.

You helped us to get the confusion wiped away, get clear on that.

But also to understand the future of how it can help us with our career, where imaging professionals need to be involved as AI continues to grow.

Here's the question that lives under everything we've talked about today.

A lot of radiologic technologists are worried.

You've cleared up a lot of stuff.

But some are still going to say I'm still worried about AI replacing us.

I don't care what they just said.

I'm not talking about the general anxiety of change is scary.

You and I talked about this before.

I remember when we went from dark room, no packs to you know, packs and people like, oh no, we need film.

How are they going to be able to see they're taking away jobs from people in the file room because you used to go hang the film.

We've seen changes over the course of the 20 plus years we've been in this profession.

Some of them have scared us.

AI is changing so fast.

But I look at now some of the things that have come from it.

Informatics.

Usually nurses only did informatics.

I have a friend that's a director over imaging informatics.

All for being a radiologic technologist because like you said, we need to be in those spaces.

One last Talking about the fear of AI, but I mean the real specific fear about job security, but about whether the profession we trained for, the career we built is going to exist in 10 years.

Is this fear valid?

Is the fear valid?

I think maybe just to draw a comparison and then I'll get a little more specifically into it.

And there's, there's so many different examples that you could, you could draw from.

And they thought, oh well, we're out of a job, the ATM can do all that.

Well, today there's more bank tellers than there were then.

What happened?

Well, they shifted to areas again where only humans can, can do and only humans can excel.

They help customers open up new accounts.

They, you know, customer relations.

They help them, you know, they answer their questions and help solve their, their problems.

All of that is very, very meaningful.

And in fact, more banks were open, so we need more tellers.

And as you kind of already alluded to in radiology and in radiography specifically.

Yeah, we go from film to CR to doctor every step along the way.

I mean, did we ever have an issue where techs couldn't get jobs?

I don't, you know, I wouldn't, I wouldn't.

And it's not, again, it's not getting better.

If there's ever been a demand for technologists, it's today when the tech, obviously the technology is as far as it's come.

And I guess just to add on to that the point that I can't speak for all the vendors, but certainly agfa, our mission is very aligned with patient care, patient outcomes.

And so that's one reason why we've kind of excelled in historically is our image processing, because that matters to patients and patient outcomes, that you have, you know, optimal image quality.

But we almost feel a responsibility and a duty to where if there's something we can do to help patients and help accessibility to care, then we're going to do that.

And if we're helping patients, then the sales, everything else kind of follows.

And so that is our vision.

That's how we're aligned.

Okay.

Yeah, Part of that is how can we help patients?

Well, we need to make sure there's enough techs to help them.

So if we can help them be more efficient and extend their capacity, then that's how we will help.

But again, it's just, there's just no, there's just no future I can see outside of the Terminator coming.

There's different types of AI.

There's narrow AI, there's general AI, there's super AI.

Everything that exists today is narrow AI, meaning it's designed for a specific intent, a specific purpose, useful features, but a specific purpose.

But you would at least require some type of general AI, which again, does not exist.

And we're not even close to how is that going to position a patient outside of maybe just a routine chest X ray.

It's just.

It's not fathomable.

It's not fathomable.

And there's so many parts that go into being a radiographer, even outside of just position, although that's the hardest, you know.

But so, yeah, I mean, is your job safe?

Absolutely.

And maybe even to go back to what I was talking about at first, why do you go into this profession?

I think a lot of us, it's because we care about patients.

That's going to matter just as much in the future, of course, and because we like technology.

And AI is almost like technology on steroids, I suppose.

So it's not taking anyone's job.

It's just trying to make a dent in this workforce shortage so that we can care for patients and continue to give them the level of care that they deserve and that they need.

Throughout the conversation, you helped those that maybe weren't convinced in the beginning or were just confused about what it was and how agfloradiology solutions does their part in helping humanize and also help patients as well with AI technology.

Practical guidance, like what is.

What are some things that you would sum up for us that technologists should start doing today to prepare?

There's a number of.

I mean, there's a number of courses online you could take about AI.

If you want to learn more about this, we do give a free CE approved webinar on AI and radiography, which goes through some of what we talked about today.

But, you know, it helps give an understanding of, of course, what is AI, what is not AI.

We give examples of AI in radiography today, and just as important, we talk about some challenges and concerns with AI integration into our departments.

You know that because I think it's important to know, like when to trust, when to not trust AI.

There's lots of information on AI out there right now.

Right.

Right now.

Right.

That you can.

That you can find.

And if you're interested in a webinar, I'd be happy to give you your department.

We have signups where you can Sign up with a number of other people that you might not know.

But we can also, if you have a department, your team wants a dedicated session, I'd be happy to do that as well.

You guys, look at all these practical, actionable things that I'm telling this podcast.

We like to bring people who are going to give you information that is honest and true, but also some actionable things that we can do from a practical level.

And that was really one.

And I'll make sure all those links, once I get them, are in the description.

But last question, and I think this is one of the most practical ones for technologies.

When should a radiology technologist trust an AI recommendation?

And when should they push back and say, I don't care what the algorithm says, Something isn't right here.

Should you trust AI?

It's not necessarily a yes or no.

It should be calibrated.

I mean, again, as we discussed, AI absolutely has the potential to extend tech capacity to help improve patient care, improve patient outcomes.

In fact, it's.

It's almost certain that it will.

That said, no technology is infallible, whether it's AEC or whether it's AI.

AI performance still relies on human programmers to some degree.

And the quality of the training data, what do I mean by that?

Well, quality can kind of be defined in a few ways when you talk about the training data used to train these systems.

One, does the data that you use to train an AI, does it represent the patient population that it'll be serving to use, like fracture detection software?

As an example, say you took X rays from just an elderly population and you tried to apply that to pediatric population?

Well, there's a decent chance you, maybe you overestimate the number of fractures or it leads to a downstream misdiagnosis.

Of course, quality data means that the data also has to be accurate.

Seems obvious, but again, to use a fracture detection example, if you have a radiologist who labels an X ray as not having a fracture, when it does and that's included in the training data, then again, that can lead to a downstream misdiagnosis data drift.

So data drift refers to the inevitable changes that occur over time, whether that's in people, disease, image quality.

Inevitably in 10 years, Covid, for example, is going to have a different presentation on chest X rays than it does today.

Image quality is only going to get better.

These are just examples of why you need to train and retrain a lot of these AI systems so they can remain accurate.

Right?

Because if you don't do that, then it can Create what's called edge cases.

So these are cases where basically they're examples that are on the edge or even outside of the training data that used maybe just to answer kind of high level, when to trust, when not to trust, when to trust.

I mean, yeah, if you have a normal patient that's in its, the AI is giving recommendations that aligns with your training, aligns with your education, then yes, you would trust that.

But also as radiographers and as professionals, we should always be thinking critically about these outputs and these recommendations.

So.

But maybe to give an example of when to question.

Well, maybe say you have a patient in the icu.

Again, challenging patient.

You're doing a portable chest X ray and the reason is to confirm tube positioning and maybe rule out pneumothorax.

Right.

You know, it's a very large patient.

So you max out your technique, you take the X ray, you look at it again, you're like, oh boy, that's, that's not a great image.

I don't know how it's the best I could do.

I don't know how the radiologist is going to tell if there's a pneumothorax or not.

But.

And you know, say on this system you have onboard pathology detection too, which exists and it's emerging and it doesn't flag a pneumothorax.

So you have a couple options.

You could say, that's fine, and you send it, you go on about your day, or you could think, oh, maybe, maybe the, you know, this, this AI pathology detection was trained on X rays from an outpatient clinic with an overhead X ray room where you have walkie talkie normal patients.

And maybe it's not.

This is outside, it's an edge case.

It's outside the threshold of image quality for it to identify the pneumothorax.

So maybe I should still call the orient provider or the radiologist and say, hey, it didn't flag anything, but you might want to look at this one sooner than later.

Could be, you know, one example and maybe to even build on that, you know, there is a real risk of what's called automation bias, which we already see today, like with the aec, for example, example where we notice techs start to lose their, their ability to set manual techniques because the aec, what we don't want is for that to happen increasingly with AI features as well.

We'll say where they are just trusting these, these recommendations without any critical thought to them and to build on that AI.

I think if you don't take anything else from this.

AI should be used to assist our clinical judgment, not to replace it.

This isn't also to say that the I don't want people to overly worry about AI is not accurate.

It's very accurate.

And in fact the FDA requires vendors to document their training data, their their labeling methodology and performance across demographic subgroups.

So there's real oversight and AI will improve care.

But again, we need to always be thinking critically about the outputs and the recommendations it's giving the same as we do today with rule based features.

I was writing down notes as you were talking because I feel like when to question, when to trust is something I get asked a lot.

It's a good question to ask, but to show how they are trained and the importance of having radiologic technologists in those rooms training those AI algorithms.

But to also to help us as technologists understand the lengths that are going to to make sure the AI technology is being used and tested and trained properly.

But also that we as technologists still are needed and need to do our jobs.

Yes, when it comes to setting techniques, don't think you can just rely on AI.

We are still technologists.

We still need to know how to do those things.

When it comes to pathology understanding that we are still responsible for looking to make sure.

I just feel like you really gave us a lot to think about and helped us with those fears and confusion through this conversation.

Jordan, thank you, thank you, thank you AGFL Radiology Solutions.

Thank you so much for coming on.

A couple of Rad Techs.

I want you to share anything that maybe you feel our audience should end this great conversation with and where they can find you and learn more about Aqua Radiology Solutions.

Some closing thoughts.

I mean if you're interested in AI, we'd be more than happy to give it.

Give you.

It's a free webinar.

I'm on LinkedIn if you want to reach out.

If you have questions based on what I talked about in this podcast, I love talking with, you know, whether you're a tech, a manager, you have questions about AI or just AGFA in general, you can reach out.

Shoot me a message there.

Hopefully this was a useful conversation.

Hopefully educational.

That's a wrap on another episode of a couple of Rad Techs podcasts.

Huge thank you to AGFA Radiology Solutions and Jordan Hermiller.

Go connect with them today.

All the information will be in the description.

It's exactly the kind of thing that's building the next generation of leaders in this field and educating us as well.

Here's what I want you to walk away with today.

The conversation we just had.

This is the kind of thing you need to be in the room for.

Not just listening to a podcast about it, but actually in rooms, at conferences, at tables where these decisions are made.

We are radiologic technologists.

AI is not going to ask for your opinion, but you can be someone who gets asked.

If this episode hit different, share it.

Forward it to another radiologic technologist who may be confused or maybe scared like you.

Because sometimes the thing someone needs to hear is just you are not replaceable.

You are adaptable.

And if you're ready to start building that future version of yourself, the one who's more than the modality, come find me.

Links for Jordan and Aqua Radiology solutions are all in the show Notes.

I'm Sean.

You've been listening to a couple of rad texts.

Now go out there and be more than your modality.

See you next time.