In this episode of Your AI Injection, host Deep Dhillon explores the transformative potential of AI in occupational therapy with experts Karen Jacobs and Alyson Stover. Karen and Alison discuss the challenges facing occupational therapy, including practitioner shortages and long wait times, and how AI-powered solutions can extend OT services. They explain how computer vision and augmented reality technologies gather objective data on patients, personalizing therapy and improving communication between therapists and families. The episode also touches on the broader implications of AI in healthcare, the importance of AI literacy for OT practitioners, and the future prospects of AI in rehabilitative sciences, and how AI tools like Korro AI are revolutionizing pediatric care through objective assessment and enhanced documentation.
Learn more about Karen here: https://www.linkedin.com/in/karenjacobsot/
and Alyson here: https://www.linkedin.com/in/alyson-stover-8a446159/
Learn more about AI in healthcare:
xyonix solutions
At Xyonix, we enhance your AI-powered health solutions by designing custom AI models for accurate predictions, driving an AI-led transformation, and enabling new levels of operational efficiency and innovation. Learn more about Xyonix's Virtual Concierge Solution, the best way to enhance your customers' satisfaction.
Listen on your preferred platform here.
[Automated Transcript]
Deep: Hello, I'm deep Dhillon, your host. And today on your AI injection, we'll be exploring the potential of AI and occupational therapy with two expert guests. We've got Karen Jacobs, a clinical professor at Boston university and Alison Stover, a leader in AI initiatives for equitable healthcare at the American occupational therapy association.
Thank you so much for coming in.
CHECK OUT SOME OF OUR POPULAR PODCAST EPISODES:
Karen: Uh, thanks for having us. We're a big follower of your podcast and, uh, it's thrilling to be on it. So thank you.
Deep: Awesome. Yeah, I'm so glad you guys are here because why don't we get started with sort of a sort of a simple question just maybe start us off by one of you telling us what is occupational therapy.
We've actually had a number of episodes on it, but I never know who's watched what or listen to what. So if you can just start off with that and then maybe tell us like what are some of the major challenges that are facing the occupational therapy community today and what. And this is the question I like to ask a lot, and what are like the inefficient realities of occupational therapy today that you envision AI, uh, improving over time?
Xyonix customers:
Karen: So occupational therapy is a health profession that works with people of all ages, people with challenges and people who are at risk, and we support them in doing their everyday activities. So Deepa, I'm going to ask you, what's really important in your life?
Deep: What do I love to do? Uh, well, I, I mountain bike and snowboard a lot.
I play a lot of guitar. Uh, I work, uh, in front of a computer. So I have some, you know, the usual array of issues with all of those three activities, which is typically at my age is like sore neck, sore back.
Karen: I, I ask you that question because An occupational therapy practitioner can be there to support you in doing those meaningful, what we call occupations.
So you already talked about, you know, sitting in front of the computer and maybe having neck, shoulder issues. Well, we could come and do an analysis of your computer workstation and provide you with strategies that you can use to be able to stay healthy. We all want to stay healthy.
Deep: So, how, how would you describe the difference between a physical therapist and occupational therapist?
Maybe that would help. So, at PT is maybe working more focused on a particular injury and the recovery of that. And you're working more on the context and environmental improvements. Is that
Karen: right? I'd like, I'd like to say that we look at the, the whole person, everything about the person. So, it may be the physical aspect, the cognitive aspect, the behavioral aspect, and it's everything that involves the person.
So in physical therapy, they really are much more focused on the physical aspect of, um, what that injury might be.
Alyson: It's the idea of just even our names, right? So physical therapy is really going to immerse itself in that biomechanical aspect of how you do things. Occupational therapy is going to pay attention to what it is you're doing and how to really modify and engage that thing that you need to do.
So, so, whereas physical therapy is most likely going to work with your body, occupational therapy may never even touch you, but still be able to find a way to change the environment or how you do an activity in order for you to continue participation.
Deep: Got it. So like when we think of ergonomics and at the desktop, for example, for somebody in the office, it all fall underneath the category of occupational therapy.
Whereas if I blow out my ACL and I need to, you know, get my, do all my exercises to get post surgery recovery, that would be PT.
Karen: Yeah, I think those are two, two examples, and I'm an ergonomist besides being an occupational therapist. So after this episode, you'll have to send me pictures of your computer workstation.
Deep: Yeah, I can talk about that forever. Give
Karen: you some suggestions. Yeah,
Deep: for any geek that makes it to my age, you usually have a whole tailored strategy.
Alyson: The big picture is that there's a That often times health requires us to work together, so a lot of times it can look like there's some sort of gray or overlap to the untrained eye, but really, I mean, we are professions that complement one another and have for many years.
There are physical
Karen: therapists, um, that do ergonomics too, uh, as well as other professionals. So that example is perfect for what Allison is saying here.
Deep: Got it. You know, what are some of the challenges that are facing the OT community today and why are we on an AI podcast talking about it? So like, what are some inefficiencies?
In how an occupational therapist normally goes about their work and how is going to help, you know, extend the reach of that O. T. or improve it somehow.
Alyson: So, Deep, I think that you might find, and your listeners might find, that probably our biggest number one challenge is that we continue to represent a very large projected growth, um, according to the U.
S. labor statistics. We're currently looking at about a 27 to 32 percent projected growth over the next 5 to 10 years. Are
Deep: you talking about population? Like what's growing? No,
Alyson: in the need of occupational therapy practitioners. However, we don't have increasing numbers of individuals who are entering into the profession.
So what we've found for a number of years now, decades really, is that the community as a whole understands and recognizes the value of occupational therapy services in their life. We're having more and more people know what it is that we can offer in your whole journey towards health and wellness. But at the same time, we don't have more people understanding what it is to become an occupational therapy practitioner.
And so We continue to experience an ever widening gap in the number of us needed and the number of us actually entering in.
Deep: And
Alyson: so then Why
Deep: is that? Is that money? Like, there's just not enough salary on the other end? Like, what are the causes of that?
Alyson: I, you know, I think that there's, I think it's multifaceted, but I would say that the number one cause is this incredible lack of knowledge of what it is that we do.
When you go into your high school, um, guidance counselor's lecture and they're talking about jobs and careers, or you go to a career fair, We're not represented in those conversations. And part of that again is like this cycle, right? I have to treat patients. I don't have time to get to the career fair to tell everybody about what it is that I do and why they should enter this profession or else I'm going to have more clients that aren't being seen.
So we're, we're in a state of sort of triaging where we can be to spread the word about ourselves. And right now it's really. In client care, and so we continue this cycle of people know about us when they need us and more people need us and are learning that they need us, but less people are learning about what it really is to become one of us and how you can enter into that career trajectory.
Challenges, right? I, uh, so I own a private pediatric practice in rural northwestern Pennsylvania. Yes, challenges are reimbursement. I don't think you're going to find a healthcare professional that's going to argue, maybe other than, um, some very specialized surgical types of healthcare professionals, but reimbursement is a challenge.
We have challenges coming from, uh, Medicaid and Medicare. Do they reimburse? How much do they reimburse? What does it look like to be a practitioner in one state versus another state with these largely state organized and operated reimbursement systems? How do people access us? Well, in most states you get a referral from a physician.
So does the physician know about us in order to refer to us? That whole piece of who do we bill How do we bill? Arguing against the various reasons why they are denying coverage, being able to do a tremendous amount of intervention, but not necessarily by the payer sources being recognized as, as being someone who can be reimbursed for that intervention.
We, as, as Karen mentioned, we take a look at your behaviors and Those are a really important piece. We also take a look at your emotional wellness and experience. We can vastly practice in a very behavioral health space, but Since we are in a country that has a health care system that divvies out that you are either a behavioral health practitioner that bills under one system, or you're a physical medicine practitioner that bills under another system, there's no place for us to bill for our full scope.
So we're forced to pick one or the other or else be a cost to the very places that we're servicing. If you look at the research, we, there was research done just prior to 2016. And in that article, not published by occupational therapy professionals, they actually found that when looking at individuals who went into acute care hospital for things like cardiac episodes, heart attack, stroke, that type of thing, the more money spent in acute care for those conditions, resulted in the only profession that reduced readmission rates.
So when you invest in occupational therapy in the acute care setting, you actually reduce hospital readmissions. We are a very effective profession, but In order to be effective, we have to practice our full scope. And we can't build for our full scope because it spans two worlds.
Deep: So, so does that come down to proof of efficacy?
Is, are part of your challenges that insurance companies just don't have kind of clear guidelines around when occupational therapies are covered or not covered or it's too narrowly covered? And it sounds like you're also, like, not all physicians know when they can Recommend to an occupational therapist and when not, so it sounds like there's a bit of an educational problem.
Just getting the word out on what occupational therapists do. But it, it almost feels also, like, related to that is, you know, like, maybe efficacy studies is that. Also a challenge and is that part of why we're talking about. Things like using video to, like, quantify behaviors and be able to, you know, maybe make cases that.
Help. Get the word out on on the efficacy of these therapies.
Karen: No, yes, and that's a great segue into artificial intelligence, because at the beginning, Allison was giving you know statistics about the need for occupational therapy, but that we have a shortage, and we're going to focus on pediatrics, and we have enormous numbers of children in the United States that need occupational therapy services.
However, there are not enough occupational therapy practitioners there to provide this service. There's wait lists for children, and wait lists can be months and months. And here's a child waiting when they could have occupational therapy. Um, right away, helping them, um, be able to address some of the challenges they have.
So here's where artificial intelligence comes in.
Deep: So before you dive into that, I, what I'd love to hear is give us an example that's representative of a normal provider patient interaction, like what is the room look like here? Okay. Who's there? What usually happens? Is it videotaped? Is it not videotaped?
Are you looking at stuff synchronously or asynchronously? Like, what are you trying to assess? Like, what kind of problem, like, what kind of problem is a child being brought into an occupational therapist room for? Um, are they even in a room? Are you going to their house? Like, so maybe give us like a really specific example of a person, of a, of a child with a case and what that interaction context is like.
Karen: Sure, and I'm going to turn it over to Allison because she's actually probably at her clinic right now. Yes, so I
Alyson: first do have to preface it with It looks different, right? Because we are very much embedded in what you need to do, what you want to do, and what you're expected to do. Sometimes a session with a client and myself looks like being able to place your order, find the right, correct amount of money, or remember the PIN for your card.
And we're in Starbucks, and we're ordering it, and we're remembering our PIN on our card. We're good. Or, we are, uh, figuring out what that idea of tipping is and then we're going into a seat in Starbucks and we're having a conversation and we're, uh, avoiding some of the stimuli that's happening around us so that we can engage in that social experience as we would with a friend.
But, but for pediatrics it largely happens either in a clinic setting or in the school setting. And a session looks like coming in, caregivers are involved, they'll enter the clinic, um, there is a room that we go into, and in that room I'm doing some interviewing with the caregivers as well as with the child.
I'm just watching that child interact, play, I'm looking at what are the toys that they're choosing. What are the activities that they're choosing to do? Uh, how long are they sitting? What does their body look like when they're sitting? Generally encompasses things like looking at how they do some writing tasks or some of that fine motor tasks.
Can they button their shirt? Can they put their shoes on? Can they take their socks off? Then it also looks at things like feeding? Are they eating a varied diet? Are they overstuffing or putting too much food in their mouth at one time?
Deep: Overwhelmed by data and unsure if AI is a risk or a resource?
Consult with our data scientists at xyonix. com and let's explore AI's profound potential together. I want to get a little bit more context. So before this patient arrived with their caregiver into your office, What happened between them and their physician to get them to send them to you? Uh, that's kind of my first question.
And then my second question is, you're having to come up with a list of scenarios to assess. Walk us through what's the process by which you select those scenarios and determine them. Like, do you get something from the physician that says, Hey, I suspect ADHD something, some, some behavioral issue in this child, and then you sort of, you know, draw on a bank of scenarios that are going to help you assess?
Like, Yeah, that would be the kind of two, the two that I would throw in right now.
Alyson: Yeah, so, I mean, in pediatrics, um, the, the, most physicians recognize that we are specialists in development. And so, they are really required to do these, checks annually or, you know, prior to the age of three, more than annually to look at how a child is progressing in their development.
So, uh, if a child has any kind of representation of a developmental delay, then they're going to refer to us. If the child is having trouble eating, gaining weight, then they're probably gonna refer to us. If the child has behaviors, they're gonna look and say, is there something sensory going on? And they're gonna refer to us.
Any kind of just issue that that red flag that would arise in their developmental checklist is going to trigger an occupational therapy evaluation referral. Most referrals from the physician just say, OT aval and treat. Say, go ahead, you do the evaluation and, and determine treatment because we are autonomous practitioners.
You don't
Deep: get suspicions or anything, like we suspect X, Y, or Z.
Alyson: No. You might get a characterization
Deep: of the symptoms though. No,
Alyson: usually not even that. Just
Deep: the markers not being met. That's nothing?
Alyson: We, no, very rarely. It's just a script that says OT eval and treat. What we do though is a phone intake. To schedule the appointment.
So, um, the physicians here, and it, it varies from place to place. Sometimes a parent is to call and say, I'd like to schedule an OT evaluation. I have a script for my physician. We have a lot of physicians just fax. The information immediately to us and then we call in that intake. We ask a lot of very daily life questions.
So, um, the first question we ask is walk us through a day with your child. What is difficult? What is easy? And that starts to begin the process of us determining what exactly is it that we're, we're looking at, um, then we'll ask some more specific questions related to the response that we get to what's the day in the life of you and your child look like
Deep: we will ask.
Isn't that weird that a physician recommends a child be seen by an OT but gives you zero information about why they're telling them to see you? That feels odd to me. It seems like at a minimum they would tell you, What, what they're suspecting or what markers aren't mad or like, you know, at this age, they're supposed to be able to do X, Y, and Z in there.
And I think they can't.
Alyson: So I think it goes back to, again, that, that space of what, where Karen says, we're really spanning across that behavioral health and physical medicine piece. They don't, they refer to OT most often when they don't know what it is. They know that there is an impairment. They know that there's a developmental delay.
They know there's something the child is not doing. So it's not something like, the child has a club foot, so we're gonna need to do the process in which, uh, the protocol addresses maybe strength, range of motion, how they walk, what they're walking in, right? To, to the physical therapist, they might be able to give that type of feedback.
But to the occupational therapist, they're just saying, I don't know, you figure it out. And so, we're really coming in as investigators. We're coming in saying, there's, there's a delay. Now sometimes they will tell us if there's an existing diagnosis. For instance, most often, about 90 95 percent of children with an autism diagnosis are referred to occupational therapy.
There's many reasons why that child with an autism diagnosis may be referred to us. They won't, they might not put that reason why, but they'll put the diagnosis, medical diagnosis of autism. Sometimes they don't put any medical diagnosis. Sometimes they write developmental delay, but, but they're coming to us saying this child can't participate.
And, and that's what, that's sort of what the golden is in OT. I don't do anything different between a child with autism and say a child with down syndrome. If they have the same delays, there's nothing that their diagnosis is telling me, there's nothing that their red flag is telling me, other than where the impairment is.
I need to look at how they're functioning, in their environment, doing an activity to determine why that impairment exists.
Deep: Got it. So when you, when, when you get a patient and the caregiver in the room, and you have to determine which activities to present to them. What's, what is your process for determining what those activities are?
Are you, is it like a function of age? And if you have an impairment that you, that you know of that you're drawing on a bank of a priori, like knowledge to like, say, okay, well, we do these things or is it something else? And
Alyson: yeah, so it, it largely comes from the intake interview with the parent. What is it that you're having challenges with look in the grand scheme of things.
All the occupational therapy professional really cares about is making it so that function and participation is meaningful and simple for you. So for instance, we can work on something like toilet training, but if you have a six year old that It's not yet toilet trained, but that is the least of your worries.
What you really need is for him to be able to put his shoes on by himself. Then that's what I'm going to work on. So it's largely determined in that interview as what are the frustration points for the family? What are the frustration points for the school, the teacher, the classroom? What are the frustration points for the child?
That's where we start. We also take a look at age, um, so that can also You know, determine where they should be. You have a parent coming in with a, an 18 month old saying, Oh, this child needs to be toilet trained. We might back up and say, well, developmentally toilet training isn't something that we really work on just yet.
Right? So we do put in that developmental piece and then with the child, we put them in a room, we have a large gym, we have ball pits and trampolines and climbing walls and swings. And, and toys. And we let the child explore. And where that child goes to gives me an indication as to what is that child's preference.
I'm going to watch them do what they prefer. What they want to do. I'm going to look at where their frustrations are coming from. And I'm going to say this is what I need to assess. We do have standardized formal assessments that we can do and that we often do, but a bulk of our work comes from observing how the child does a specific activity and where the frustration points are and where the ease points are.
We're never going to tell you. So for instance, a lot of times people will say, Well, an occupational therapy professional will work on handwriting, and so obviously they look at pencil grip. Well, the fact of the matter is, is I might look at your pencil grip, but if you're not complaining that your hand hurts and your writing's legible, I actually don't care how you hold your pencil.
It really doesn't matter because it's not affecting you in any way. Now, you could have a perfect pencil grasp, but Say you get tired after 10 seconds of writing, or with your perfect pencil grasp, you have really illegible writing. Well, now I'm going to look at your pencil grasp, and I'm actually going to take it away from that perfect pencil grasp, because that's not functional.
Deep: So, part of what I'm hearing is there's a room, a physical space, there are activities. There, you're taking in a lot of your information by what you observe and visually see, and I'm, I've got a note here that, you know, we opened up this conversation with there's not enough occupational therapists and not enough folks going into it.
What are we talking about when it comes to AI? Are we talking about video cameras, making this a less synchronous activity so that we can rely on occupational therapists to train a model? To figure out behaviors in these kinds of, uh, contexts and environments. So that an OT doesn't have to physically be in the room all the time.
And maybe we can, like, learn from that. Like, what, what are we kind of thinking?
Karen: So I'm glad you asked that question because part of what Alyson was sharing was something that's in actually physically in that space. But we've been involved in telehealth also.
Occupational therapy through virtual context. We didn't want to shut down our clinics. Uh, Allison's Clinic Capable Kids turned to using telehealth technologies. Zoom, maybe it was a telephone. Whatever it was to continue to provide the services. So telehealth, um, really is something that, Because of COVID, jettisoned occupational therapy in using it.
And it was something that we had wanted to use. Some of us were using it. I had been using it in the area of ergonomics to do computer workstation evaluations. You know, with a camera, I could see everything. I could, you know, um, have things measured, all of that. But it allowed us then to begin to look in the natural environment virtually.
of the children and other clients. And so that's one, um, example of what, you know, we were able to do. Virtual reality and simulations is another thing with AI. But I want to start by saying, we all need to become literate in artificial intelligence. And that's why I love your your podcast so much. Um, it has been a mission for me to help occupational therapy practitioners and students to not be frightened of AI, to use it in a responsible and ethical manner.
And so I've started this sort of momentum with others of, we're calling it the, the OT AI revolution. Where we want our community of practice to be educated. So now, those points that we were talking about, where there are issues about not enough occupational therapy practitioners, and I'll stay with pediatrics and lots of children that, that need us.
About, it's over a year ago, that a colleague contacted me and said, That she had been working with a company that was creating an AI tool, and that's called Coro AI. And what she liked about it, and what made me feel inspired to actually have a conversation with the owner of the company, was that they were having occupational therapy practitioners working with these data scientists.
individuals, 30 of them that are top ones in designing the algorithm with CoroAI, they're called gaming experiences that look at the skills that we as occupational therapy practitioners would look at in evaluating and working with a child. Because of my role as the Associate Dean of Digital Learning and Innovation at Boston University Sargent College, I said, of course, I want to learn about the new technologies and AI is beyond chat GPT.
So I met with the owner. He flew into the Boston area. And by the time we finished, I said, I want to be part of this. Because what this product was bringing was the objective data. Um, That has been so challenging for us to obtain.
Deep: Overwhelmed by data and unsure if AI is a risk or a resource? Consult with our data scientists at xyonix.
com and let's explore AI's profound potential together.
Karen: Allison could be, as she was describing, looking at a child and saying, oh, bilateral coordination seems to be one of the challenges for this child, or some fine motor coordination may be a challenge. So here's where The Coro app comes in. They're using computer vision.
So my, uh, the cell phone is here, uh, looking at the child.
Deep: The caregiver is, uh, is manning an app to, uh, to capture the imagery.
Karen: So the images are not captured. We're seeing them in real time. There's no videos going on. So, yes. When the app is paired with the device at home, and it could be a smartphone, it could be a tablet, the caregiver is able to work with the child and help them set up.
So what would happen, I'm pretending this is the, the phone. Okay, and it's looking at you and it is looking, um, at you without sensors and, and being able to say, yes, you are, um, you are looking at me. The app. And I am seeing that you are doing bilateral coordination. Maybe we're climbing. And I'm recording that, what I see.
And then the child stops the mission that's part of the, the gaming process. And then the occupational therapy practitioner sees immediate data. And what's
Deep: the data? It's biomechanics information. Like what exactly is the data? So you, you, you got somebody holding a camera, pointing it at a kid, a kid's doing an activity.
I've so far I in my view I've got I've got video but how did I get from video to Yes, it's different information you're talking about.
Karen: Some of the things that we need to look at are different skills, like bilateral coordination, visual motor integration, the ability impulse control. Um, so all of that is being registered in the child, um, actually doing the game.
Is it a physical
Deep: world or a virtual world? And who's giving the data? Is it the machine giving the data or the human? The algorithm
Karen: is giving us the data.
Deep: Okay.
Karen: So, um, we're in occupational therapy and how we've done this is we don't expect this to replace what the occupational therapy practitioner is doing, but we ask for them to introduce it for maybe the first five or 10 minutes in a therapy session.
So the therapist would set up the smartphone. Or a tablet and, and, um, come up with the levels based on what they have seen the child be able to do. And the child, um, is playing a game, so they're climbing. They are actually seeing themselves on the screen itself, too. While they are moving as the astronaut.
And they may be avoiding some circles, or they may be trying to get minerals for their mission. Some kind
Deep: of augmented reality experience?
Karen: Some, some of that, yes. It's immersive in the sense of that they become that astronaut.
Deep: There's a character on the phone, they're physically climbing, they can see the phone and the character on the phone.
And so there's an exercise that's basically been kind of canned. And through the process of them interacting with this augmented, I guess, virtual representation of themselves, you're somehow able to glean data. About their cognitive ability and these other abilities is that all
Karen: these abilities. Yes. All of these skills were able to get and what I think is exciting is that the Computer vision will actually correct the child.
So we may want them to be re, take, take a look at how I'm reaching. So I, they may want me to reach like this or cross my body like this in the activity I'm doing. And if I'm not doing it correctly, um, the app will correct the child. It's like highly motivating. And so what we've been doing in our research, um, because I've been conducting research at Capable Kids, is looking at the feasibility of using this.
So then we can go back in time and say, we have burnout in our population, particularly with occupational therapy practitioners seeing many, many children, one right after the other. We are. burdened, and I hate to use that word, with documentation as well. And so what we've done with, um, how to use Quoro is we suggest that the occupational therapy practitioner start off their session with five to ten minutes of Quoro, engage the child, um, let them do their missions, introduce this to them, get that objective data that we can use for documentation, and then Continue with our regular occupational therapy practice
Deep: longer term.
Your goal would be to not only get the data from an occupational supervised session, but maybe have the app with the caregiver be able to get to additional data to.
Karen: Yeah, so you just talked about the experimental design of the research project. So we have a control group and experiment group and the experimental group.
We paired the parent's phone or caregiver's phone, um, at home so that the child could continue to use the app at home. And what's exciting about this, um, as an occupational therapy practitioner, when we work with people and we send them with homework. Because we want them to keep doing the skills that we were doing in occupational therapy with them.
We find they don't want to do their homework. What we're finding with Quoro is the children are so motivated, um, to do their missions, um, that they are continuing to do the skills that we were doing in occupational therapy with them. And so it's very reinforcing of what we're doing. And over time, perhaps we'll reach the goals that we've set for the child sooner.
Deep: Let me see if I can summarize this. So you're able to leverage a real occupational therapist and a patient with a caregiver to gather more objective data in an augmented reality like experience where you can kind of design your exercises through this application. Um, that will facilitate gathering of that data.
That data will help you ultimately, like, understand, presumably, they're also getting training data from the OTs with any kind of maybe not diagnostics only, but like assessments. And then the idea that was ultimately to get the machine to be able to, like, score the students on these, uh, specific capabilities that the OT would normally give, and then be able to maybe flag potential issues.
To sort of extend the reach of the occupational therapist, something like that.
Karen: Yeah, and it is and it's it, but that's that's very simple. It's even it's even bigger than than that. Um, it's providing. Objective data for us, um, and data that we haven't been able to to do so long term. I think it's going to help us with a long wait list that we have, because we can have children before they come into occupational therapy, working with the coral app.
And some of those gaming experiences, and after they finish occupational therapy, because, you know, we are limited by reimbursement. They could be. With their family, um, using the coro AI at home, which is reinforcing the skills that they were developing in occupational therapy.
Deep: Sorry, but do you think there's a, so I've seen other occupational therapy applications, like early autism spectrum disorder detection and, um, like.
Tracking of early markers and infants. Um, this is a project that we actually worked on with, uh, with a company here in Seattle called early markers.
Karen: I know, Teresa.
Deep: Okay, great. Terrific. So there is no augmented reality context in their application. The video analysis is sort of directly going to assess sort of behavioral marker.
kind of objectives and then, you know, identifying them. Do you think there's a, a challenge or a potential problem in putting the virtual, the augmented reality thing in the middle of the kid? So now that the child has to understand this device and it's like behaviors that are sort of done in the context of a device, as opposed to.
Like, you know, no device other than capture.
Karen: Yeah, so our expectation is not that the child would use this all the time. Um, our expectation is that it would reinforce some of the skills that we're working on in occupational therapy. These experiential games are exhausting.
Deep: Okay.
Karen: So we're not going to see a child saying, you know, um, I want to spend or spend two, three, four hours on these games.
We'll see them maybe spend 10 minutes, 15 minutes a day doing the game. So I don't think it is going to have that scenario that you were describing. Uh, the other thing that I think is But it does,
Deep: it does like de facto limit you, right? Like you can't, you can't use a tool like this for infants. Because they,
Karen: at this point, at this point, the way this tool is designed, um, it's for children ages four to 12.
However, with the technology that Coro has, they are looking at younger ages, and eventually they may look at older ages as well.
Deep: Got it.
Karen: So absolutely. Um, and, um, I, I just think that the infrastructure that they have with this algorithm has so many opportunities and why I think this company is Uh, important and why I'm an advisor to them and doing research is that I like that they're listening to us.
We really are designing partners with them. You know, we can say we're playing, we see a child playing this game, would you consider adding this feature? So they're listening to us, which is what I think is very exciting to be on the ground level. But I want to. Really look at the potential impact of OT and AI and this sort of revolution that we're trying to do.
Number one, we need to educate our occupational therapy practitioners and students. On what is AI, because I think a lot of people now think AI is Gen AI. They don't realize that it's beyond, beyond that. So one of the things that we've done at Boston University, particularly in our college, is, um, we've worked with a team, including a computer science, um, student to create online training on optimizing the intersection of AI and health and rehabilitation.
Sciences, and that will be something that our students will be using as well as training our faculty on how to use it. So I think education is the bottom, bottom line is, as we're starting with this AI revolution. We've talked a lot about the broader reach. that having AI be a partner with us, walking side by side, um, in the clinic, being there with us, it will broaden how many children in particular we can reach where with our shortage of occupational therapy practitioners and the growing number of children that need occupational therapy, this will help us as well.
Deep: Got it. So we're trying this new thing here on your AI injection, where we do this thing. We call it the fast five. So it's like five fast questions. So the first one is, how does AI help personalize occupational therapy?
Karen: It helps personalize it because it's actually getting objective data that is of that person itself.
Deep: How does AI, how can or does AI assist in early diagnoses?
Karen: With, uh, appropriate objective data, it may be able to provide, I think, confirmation of a diagnosis. As occupational therapy practitioners, we don't diagnose.
Deep: Maybe you get assistance with data gathering on particular behaviors, for example, that seems like something that I could assist with.
And beyond diagnosis, just like building a case for where someone, where a patient is at any point in time. Third one, how does AI improve communication between therapists and patients? Or how could it?
Karen: You know, um, objective documentation that we can share immediately with a family member, I think, opens communication where, uh, it takes a while to do documentation.
And with AI, it's, you know, our thought partner.
Deep: Yeah, now you can kind of integrate it into how you speak with your patients or their caregivers.
Karen: Absolutely. And, and we can have documentation very quickly.
Deep: Number four, biggest challenges when implementing AI in occupational therapy.
Karen: I think helping people understand how to use it.
I think for everybody, how to use it responsibly.
Deep: Last question. Future prospects of AI at OT.
Karen: Sky's the limit. I want to see occupational therapy practitioners being the people who are helping to design the use of
Deep: AI. Awesome. So, one question I always ask to wrap up on is, let's fast forward five or ten years.
Thank you very much. Um, our machine learning, um, AI systems continue to evolve, our computer vision, CV, all that stuff gets better and better. What does the world of occupational therapy look like, uh, in your view, if everything kind of pans out the way that you envision and would like?
Karen: I think we'll be able to have outreach much broader than we have right now.
Also, when people see that we're using artificial intelligence, it might also attract people to come into the field. And so it could help with the shortage that we're going to have with the number of occupational therapy practitioners out there. So I think it's exciting. I think it's a revolution in occupational therapy, and we haven't had a revolution in occupational therapy.
My hope is that with the Our profession, feeling comfortable with AI, becoming literate in AI, that will help change the face of health care.
Deep: Well, thanks so much for coming on. That wraps up today's episode of your AI Injection. I'm Deep Dhillon, thanking you for joining us as we continue to explore the highly nuanced world of AI.
A world full of both tremendous promise and potential pitfalls. Stay tuned for more insights and visit us at xyonix. com. That's X Y O N I X dot com to learn how we can help you implement transformative AI solutions responsibly.