YT YouTube2Text

YouTube2Text → Transcripts → Can We Trust AI in Veterinary Diagnostics? w/ Dr. Jordan Towns | The Veterinary Roundtable - EP189

Can We Trust AI in Veterinary Diagnostics? w/ Dr. Jordan Towns | The Veterinary Roundtable - EP189 — Transcript

by The Veterinary Roundtable · 17,918 words · 2,592 segments · language en · Watch on YouTube

Open in transcriber Summarize with AI Download SRT/VTT

Full transcript

0:01[music]
0:04Hello everyone and welcome to a special
0:06episode of the veterary round table. The
0:07podcast where we answer your veterary
0:09related questions while having some fun
0:10along the way. If you enjoyed today's
0:12episode, be sure to leave us a review on
0:13your podcast provider of choice. And if
0:15you have any feedback to offer to
0:16improve the veterary round table, let us
0:18know.
0:19>> All right,
0:20>> let us know. We're here for it.
0:21>> Don't have Courtney or Ally here because
0:23they're both on maternity leave and this
0:25would be the this the section where
0:26Courtney would tout our socials. So, I'm
0:29gonna throw it to Duckwall.
0:31>> Okay. [laughter]
0:32We have social channels. It's on YouTube
0:34and the other socials, Ticketytac and
0:36Instagram. Uh, it's at the Veterary
0:40Roundtable. So, we kind of we've kind of
0:43slowed down on some funny videos. We're
0:45promoting the live show, which by the
0:47time this comes out, we would be done
0:50with the live show.
0:52>> Oh my gosh. Is that true? That is true,
0:54isn't it?
0:55>> That's true.
0:55>> It is true. Our live show is this
0:57Sunday, August 16th.
0:59>> Okay, let's do like a whole Let's do a
1:02whole like pretend like we already know
1:04what happened at the live show.
1:06[laughter]
1:08>> The live show was so amazing. Oh my
1:11gosh,
1:12>> it won a lot of money.
1:14>> Yes, we met a lot of cool people,
1:17>> right? You know, so it was one big fun
1:21event.
1:22>> Can't wait to do it again next year.
1:25>> There we go. Okay.
1:26>> Soon. Okay. Great. Yeah.
1:31[laughter]
1:37>> It was insane.
1:38>> We had to go and get an epipen and it
1:41was wild.
1:42>> Yes. The bales of a were everywhere.
1:45>> You know the horse that rode in and
1:47y'all like there was like a flaming
1:49spear. Like that was pretty sick.
1:52>> Wasn't that so cool? And then King
1:54swinging from the banisters of the
1:56ceiling like what?
1:58>> It was pretty much circus Olay. Like
2:01>> that's right. When you look at me,
2:02that's what you think. Don't you think?
2:04I mean like Circus Olay. I mean,
2:06>> absolutely.
2:06>> You know, uh you know, I'm that
2:08grateful. So,
2:09>> we'll be in Vegas. Don't you worry.
2:11>> Yeah, we'll be in Vegas. That's right.
2:12At the uh Sphere.
2:14>> Oh goodness. Okay. Well, that sounds
2:16like a great time. Can't wait.
2:18>> Yes.
2:19>> Okay. So, today is another special
2:21episode. We love these episodes. We get
2:24to learn so much from knowledgeable
2:26expert level people. And so I'm going to
2:29introduce our special guest. She is an
2:32assistant clinical professor of clinical
2:34pathology at Auburn University and an
2:36Auburn alum through and through. She
2:38earned her undergraduate degree,
2:39veterary degree, residency training, and
2:41PhD at Auburn. After working for several
2:44years in general practice and emergency
2:46medicine in Birmingham, she returned to
2:48Auburn in 2017 and completed her
2:50residency in 2020 and PhD in 2024. After
2:54finishing both, she subsequently joined
2:56the faculty. She's passionate about
2:58teaching veterary students, graduate
3:00students, and residents. Her clinical
3:02pathology team provides timely
3:03diagnostic services including blood
3:05work, cytology, and ur analysis. Outside
3:08veterary medicine, she competed for team
3:10USA in the modern pentaplon from 2018
3:13through 2024.
3:16Serves as a college football official
3:18and plays the pipe organ. Please welcome
3:20Dr. Jordan Towns to the veterary round
3:22table.
3:25>> Welcome.
3:25>> Thank you all so much for having me.
3:28>> So excited.
3:29>> We were just before we started about the
3:32pentacleon and how I didn't even know
3:33how to say the word correctly. So,
3:36[laughter]
3:37When
3:38>> when Dr. Tals reached out via Instagram
3:41and then I was like, you know, stalking
3:43you, I was like, what? She did what? She
3:46does what? Oh my gosh. And she does
3:48what? I was like, we're having her on
3:49just because she does all of these
3:50things. And we have a really cool topic
3:52that we're going to talk about. But I
3:54think we got to talk about some of these
3:56personal like things that you do outside
3:58of veterary medicine because I think
4:00you're a fantastic example of veterary
4:03medicine not defining you. It is a part
4:05of your life, but there are so many
4:07other things that you do to fill your
4:09bucket. And so I think that that
4:11sometimes can be missing for a lot of
4:13veterinarians as they go out into work
4:15and into practice. And so totally,
4:18>> you know, it's it's just so cool. It's
4:19such a cool part of your story.
4:22>> Oh yeah. Um yes. I I feel like uh my
4:26goal in life has been to not leave any
4:30stone unturned. So [laughter] um so
4:33we're trying. you know, there's still
4:35goals on the horizon, but uh but we can
4:37definitely talk about the things I've
4:38done up until now. So, yeah.
4:41>> Yeah, that's fantastic. So, what in what
4:44is what I have so many questions. The
4:46pipe organ, when did you learn how to do
4:48that? Like, that's amazing.
4:51>> Okay.
4:51>> Well, Doug, well, Doug, well, I'll have
4:52her start in the beginning. Like, where
4:54Yes. Like, just lead us down your road
4:57of your life. Like, let's just talk.
4:58>> Okay, good. Okay, that's good.
5:00>> Um, yeah. Where should we start? Okay,
5:03so I guess we should start in high
5:04school because I actually did not intend
5:07to be a veterinarian. Um, I went to the
5:10School of Fine Arts in Birmingham,
5:12Alabama because my dad was a graphic
5:15artist and I wanted to be an artist. Um,
5:18and I like the School of Fine Arts, it's
5:20kind of like a magnet program, but not
5:22quite a magnet program. It's uh you know
5:24you you go to the school and then it's
5:26it's a normal high school and then you
5:28have um an extra three hours a day of
5:31your art specialy. Um so you know we had
5:34visual arts, creative writing, dance,
5:36theater, music. Um and then there was a
5:39math and science department, but I was
5:40not in the math and science department.
5:42I was I was in visual arts. So, um, but
5:45yeah, that's, uh, that's where I got my
5:47start. And I went through that whole
5:49program and realized that if I did that
5:52in college, I would essentially be doing
5:54the whole thing again.
5:56>> And [laughter]
5:57I'm like, I just, you know, I love art,
6:00but I did not like going to art school
6:02enough to do another four years of it.
6:04So, um, at that point, I knew I'd always
6:07loved animals. I loved horses the most.
6:10So, I really actually wanted to be an
6:12equin vet. um when I first started and I
6:15went into animal sciences undergrad um
6:19which like loved them to death um at
6:22Auburn. Auburn's animal sciences program
6:24is just like part of my defining uh it
6:28defined my character. I had some
6:29excellent professors there. It was
6:31wonderful. Um and uh then after that I I
6:35got into vet school. Um, but before that
6:39happened is when I was in undergrad is
6:41when I started racing cars. Um,
6:44>> oh my gosh, this is amazing. Okay, keep
6:47talking.
6:48>> Like watching a movie. We got
6:50>> I know. This is so cool. Okay,
6:53>> so um if uh so my my undergrad was like
6:562004 to 2008 and my high school was like
6:59the four years before that. So 2000 to
7:012004. Anybody's familiar with like the
7:04pop culture during that time? Fast and
7:05the Furious was a big thing.
7:08>> Um, and all my friends were into it and
7:10there was like Gran Turismo video games
7:12and all that and I just loved cars. Like
7:15I just thought they were so fun. And I
7:18remember um after I graduated high
7:21school and like I really wanted a Toyota
7:23Celica. So I actually still have it to
7:26this day. Um my little Toyota Celica
7:28with 230,000 miles on it. But [laughter]
7:31I really wanted a stick shift. I was
7:33like my parents thought I was crazy.
7:34They were like, "You're getting a stick
7:36shift. You don't even know how to drive
7:37it." And I'm like, "You're right."
7:40So anyway, um I uh one of my friends
7:43went off to another college, saw an
7:46autocross going on. So autocross is kind
7:48of like beginner car racing, but like
7:52you can go all the way the national
7:53level with it. It's it's like entry
7:56level car racing. I'll say that because
7:57it's in a parking lot.
7:59>> It there's cones, there's timing
8:01equipment. You just go around the course
8:03that they build as fast as you can.
8:05You're not really risking life and limb
8:08necessarily. You're not risking wrecking
8:10the car. So, you know, it's a it's a
8:13safe way to explore motorsports. Um my
8:16friend uh my close friend from high
8:18school had been to his college, seen one
8:20of these things, came back to
8:22Birmingham, which is where we both lived
8:24and we're both from there. And he was
8:26like, "We should build one of these
8:28autocrosses. There's one in Birmingham."
8:29And I said, "Cool, let's go." Um, so, so
8:33I showed up and I knew literally nothing
8:36about what I was doing. I just had this
8:37car and uh I was like, "Okay, what do I
8:41need to do?" And they were like, "Uh,
8:42well, you need to pay $40 and register
8:44and um, you need a number." And I was
8:47like, "Cool, give me a number." And they
8:48just gave me a number. And I was like,
8:50"Okay, sounds great." And, uh, you had
8:52to pass a tech inspection, which I
8:54almost didn't pass because apparently my
8:56battery didn't have a tie down. And so I
8:58had to fix that.
9:00Oh my god,
9:01>> it was amazing. It was amazing, let me
9:03tell you. And then um the story that my
9:05autocross friends like to tell to this
9:07day is that I pulled up to the I was so
9:10excited. I pulled up to the line, I like
9:12revved the car like up to the rev
9:14limiter and dumped the clutch and almost
9:16stalled it out [laughter]
9:18and
9:19>> and then it just went up from there. So
9:22um so yeah, I that was my first like
9:24real hobby of my own, I would say. Like
9:27you know, I'd done things before. I'd
9:28ridden horses, I'd been a runner, things
9:31like that. But like autocross was like
9:33my first like defining moment hobby. Um,
9:37and since then, I mean, I autocross to
9:40this day. I just got my 20-year pin from
9:42the Sports Car Club of America.
9:44>> Wow.
9:44>> Um, I've raced all kinds of cars. So,
9:47started with a Celica, but went on to uh
9:50Toyota MR2, Miatas,
9:53Subaru WRX,
9:55uh Mazda RX8, and then the latest is a
10:00Toyota Supra. So, um so that's our car
10:04this year. And um I have I think six
10:07national championships. So, um gosh,
10:11>> yeah, it's it's an amazing hobby. Like,
10:14it's just it's the most fun. So, anybody
10:16that wants to like get into motorsports,
10:19like check out the Sports Car Club of
10:21America, SECA Autocross, it's the way to
10:23go. It's so much fun. So,
10:27>> so that was that was my first hobby. Um,
10:29still have it to this day. Love it to
10:31death. Um, and then, uh, so the rest,
10:35you know, after that, it was kind of
10:37occupied by vet school. Vet school takes
10:40up a large chunk of your time as we all
10:42know. Um,
10:45>> and uh, and I I actually really enjoyed
10:47vet school. It was a lot of work,
10:49obviously. It's a lot of time. It's a
10:51lot of commitment, but um, but I did
10:53enjoy it. And, uh, I think, you know, I
10:57went to Auburn, like you said, I've I've
10:59got three degrees from Auburn and a
11:01residency. So, um, Auburn has been good
11:05to me. uh you know to you know kind of
11:08jokingly quote um but uh Auburn Auburn's
11:12been a great place. Um I I really got a
11:16lot out of our clinical rotations. I
11:19think that's done very well. Um that
11:21school's done very well. We have a lot
11:22of awesome professors. So and now I get
11:24to give back in that same way. Um then
11:28after that I went I worked for six years
11:32in practice. I started out in mixed
11:35animal um and then transitioned to
11:39emergency work which I thoroughly
11:42enjoyed but also it was very stressful.
11:45So I stayed in that I think three or
11:47four years. Um and then transitioned to
11:51being a relief veterinarian for the last
11:53two before I went back to residency.
11:56>> Um and then I started residency in 2017
11:59and that's when the other two hobbies
12:02kind of took off.
12:03Um I I started um my pentathlon journey
12:10at that point. Uh and the way that
12:13started was a few years before I was
12:16watching the Olympics, the 2016
12:19Olympics.
12:20And I was watching the opening
12:22ceremonies and I was like, "Wow, that
12:26looks like a lot of fun. I want to do
12:28that. I want to be in the opening
12:30ceremonies."
12:31>> [laughter]
12:32>> Oh my god. Yeah. And being as though I
12:36had not trained in a sport since I was
12:38like five years old, like I I was like,
12:42what do I do? And so I like I I just
12:45told this idea to a few of my friends
12:48and I like posted it on Facebook and was
12:50like, what sport should I do? And um
12:56why not? I I had a background in
12:58horseback riding, which unless you have
13:00like millions of dollars, like probably
13:02isn't going to work. And um I also had a
13:05background in shooting. So, um I am a
13:08pistol instructor. That's a whole
13:09another story. Um but [laughter] but I
13:12was like, you know, I could do shooting.
13:15Um that seems like something you could
13:17be any age. And you know, there's like
13:20shooters, they're in their 50s at the
13:21Olympics. So I was like, yeah, I could
13:23totally do that. And so I posted this on
13:25Facebook and one of my friends
13:26said,"Have you heard of modern
13:28pentathlon?" And I was like, "I have
13:30not." And she said,"Well, you already
13:32kind of do three of the five sports."
13:34And I'm like, "I'm listening." And so
13:36she said, "Well, the three sports you do
13:38are horseback riding, running, and
13:41shooting. And the other two sports are
13:44swimming and fencing."
13:46>> And you would only have to learn two
13:49sports.
13:51[laughter]
13:51>> There you go. as opposed to just doing
13:53one,
13:54>> right? You know, I I went in looking for
13:56one sport and I ended up with five. So,
14:00um, so I was like,
14:01>> "Sign me up. Let's go figure this out."
14:03So, I I emailed I I Googled it. I was
14:07like, "How do I get into modern
14:08pentathlon?" And somebody's name in
14:10Atlanta came up, which was the closest
14:12name. It's Keith Barry Hill in Atlanta
14:15um has a a pentathlon team, Atlanta team
14:18pentathlon. So, I emailed him and he
14:21called me and he was like, "Hey, let's
14:23talk about it." And he's like, "I have a
14:25competition coming up in six weeks. You
14:28want to come?" And I was like, "Are you
14:29kidding?" Like,
14:32>> um, I got to learn to swim first. And
14:33[laughter] he was like, "That's fine.
14:37You got this." Um, and he told me, "Go
14:40join Auburn Master Swimmers." And so I
14:43was like, "Okay, sounds great." And so I
14:44showed up to Auburn Master Swimmers and
14:46they're like, "Yeah, we'll get you set
14:47up. Here's Lang. How about it? And I was
14:50like, "No, no, don't know how to swim. I
14:53know how to not drown. That's what I
14:55know." And they're like, "It's it's
14:57fine. Just start swim." He goes, "Uh,
14:58Connor Bailey," he was helping me. He's
15:01uh kind of the leader of Auburn Master
15:02Swimmers. He's like, "Get in the lane.
15:04Swim a lap. Let's see what you got. Can
15:06you swim with your face in the water?" I
15:08said, "Yeah, [laughter] I think so." And
15:09he was like, "Okay, go." And then so I
15:11did. And he was like, "You're great. You
15:14It's fine." And I was like, "Okay, okay,
15:16sounds great." Um, so then I started
15:18swimming with Auburn Master Swimmers and
15:20they do have coaches and they helped me
15:22a lot. Um, YouTube videos also helped me
15:25a lot. [cough and clears throat] And uh,
15:27and there was at the time a fencing club
15:30in Auburn. So I joined Auburn Fencing
15:32Club. Um, and uh, the coach uh, Ryan
15:36Dilap, he's um, he's now in Birmingham,
15:39but he helped me a lot, too. I want to
15:41make sure I mention all of these names
15:43because these are people that have
15:44helped me so much along the way. Um, but
15:47anyway, I legitimately went to that
15:49competition six weeks later and and just
15:52yeah, it was just that was the beginning
15:54of it. Um, so after that, like I did go
15:58to national championships that year. I
16:00didn't do very well, but I mean, you
16:02know, I've been training for six months,
16:04so it is what it is. Um, and then I just
16:08stayed with it for six years.
16:10um until uh 2023
16:15is when I really got uh that's when I
16:18really made the level of the national
16:20team. So um then I I got to go to events
16:24um in the Dominican and in Poland and in
16:29Madrid. Uh and then the next year was
16:342024. That was the Olympic year. And so
16:37that's the year that everybody's, you
16:38know, like everybody's going for it. And
16:41I went to the um team qualification, the
16:44national team qualification event that
16:46year and I came in third. So I made the
16:49team. They take like the top six. And so
16:53I made the team as an alternate. And so
16:55I got to go to two World Cup events in
16:57Turkey and Bulgaria. Um and I also got
17:01to go I mean I got to go all over the
17:03world. It was so fun because they had
17:04other events that weren't necessarily
17:06World Cups. It was just like other
17:07points events that you could go to. Um,
17:10so yeah, I just it was amazing. Like it
17:13was an amazing experience. [laughter]
17:17>> Do you take do you take your horse with
17:19you to all the countries?
17:20>> Oh, that's a great question. Um, the
17:22horses are provided for you.
17:24>> Okay.
17:24>> And it is a random draw.
17:26>> Okay.
17:27>> So, um, yeah. So, so you get 20 minutes
17:30to get to know your horse and then you
17:32get to go jump a course, uh, a show
17:35jumping course with the horse.
17:37[laughter]
17:38>> Yeah.
17:41>> Oh my goodness.
17:42>> Yeah. Um, and that's where I really like
17:45to thank my background. I I owned a
17:48horse for a few years, but other than
17:50that, I've just ridden different horses
17:52and different lessons and stuff. And
17:54another shout out, uh, Stephanie Counts
17:56in Auburn, uh, at Silver Lioning
17:58Equestrian Center. She was so down with
18:01this. She was like, "Come out to my
18:04farm. I got I got horses. You know, you
18:07know, I've got lesson horses. We can put
18:09you on a different one every week and
18:11you know, you can have your 20 minute.
18:13We can set up like a pentathlon. You
18:14have your 20 minutes." Oh.
18:16>> And then you go jump the course. And I'm
18:17like, "Cool. Sounds great." And I that's
18:19what I've pretty much done my whole
18:21life. I've ridden all all different
18:22courses. So, I did think that was an
18:24advantage for me cuz if you like own one
18:28horse and you get to know that horse,
18:30like then you're kind of like locked
18:32into that horse. But like if you ride a
18:34whole bunch of different horses, you
18:36learn how to communicate with the horse
18:38very efficiently, very clearly.
18:42And I actually never like, knock on
18:45wood, because I still do this sport to
18:48on the mast's level, but I never had a
18:50bad horse. Um, and so I had a great time
18:54uh with the horses.
18:56>> Yeah,
18:57>> it was it was wonderful. Um, so yeah.
19:00Um, that was the horseback riding. The
19:01fencing is um I nobody's familiar with
19:06fencing, but [laughter]
19:07um but fencing uh essentially you touch
19:10the other person before they touch you
19:12and there's like electronic equipment
19:13and you get a light if you win. And so
19:16it's um in pentathlon you just get one
19:19bout against each other competitor.
19:21Okay?
19:21>> And so it's whoever wins the point you
19:23get the victory. So um it's all
19:26pointsbased. So if you got 23 victories
19:30and
19:31>> 17 defeats or whatever it was, that
19:33would be a points amount.
19:35>> And so then you won that number of
19:37points in fencing. And the same for
19:38swim. Whatever time you swim, you're not
19:40necessarily swimming against the other
19:42competitors. You're trying to get your
19:43best time. So swim and then the run is
19:46combined with the shoot. So you run and
19:49then you have to shoot and then you run
19:51and then you shoot. So um it's all about
19:54controlling heart rate, you know,
19:56accuracy after
19:58>> after running. So um so that's kind of
20:01how the sport works.
20:02>> Wow.
20:03>> So it's fair to say when you set your
20:05mind to something, you're going to do
20:06it. And then
20:08>> um how so then how does clinical
20:11pathology come from being a very like
20:15high functioning athlete competitor like
20:19go-getter? Where did you come I mean you
20:22did a lot of things in the vet field.
20:24You had your toes in some of it. So why
20:26clinical pathology?
20:27>> Okay. Clinical pathology. Um, so when I
20:31was like I would say when I was in vet
20:34school and like getting ready to finish,
20:37I knew that I wanted to specialize in
20:40something at some point. Um, and I just
20:44didn't know what it was yet. And I decid
20:47like I chose the path of I want to go
20:49work, you know, see what like I, you
20:53know, I kind of gotten tired of school.
20:54I was like, let me go work. let me see
20:56what everything's about and then if it
20:58works out in the future I'll go back and
20:59do a residency and you know working in
21:02practice would help me figure out what
21:05exactly it is that okay
21:06>> I wanted to do and so I had considered
21:10internal medicine I'd considered surgery
21:12I'd considered emergency critical care
21:15um honestly clinical pathology was not
21:17really on the radar um until I got into
21:21practice and um I really enjoyed the
21:27diagnostic aspect of everything.
21:29>> Mhm.
21:30>> So like
21:32the lab work, the microscope, you know,
21:36like cytologies
21:38and I I did love internal medicine and
21:41critical care. But the more I thought
21:43about it, the more I was like, what do I
21:45really like about I like the diagnostic
21:47aspect of it and tying that back in with
21:50my family. My mom is uh she's a medical
21:53technologist who retired um she's a
21:55human medical technologist. She retired
21:58uh you know a few years back. But I had
22:01essentially grown up in a lab um because
22:04of her.
22:04>> Okay.
22:05>> And so that's kind of you know when I
22:08thought about it I was like clinical
22:10pathology. Oh I wonder what that's like.
22:13I remember my clin path professors. I
22:16remember them talking about blood work
22:17and diagnostics. I'm going to call them.
22:19I'm gonna, you know, see what it takes
22:22to do a clin path residency.
22:24>> So, um, so the the same professors that
22:26I had worked with were still there, um,
22:29you know, when I when I called them
22:31back. Um, and so I talked to, um, Pete
22:34Christopherson and Beth Spangler, they
22:36they both worked with me and they were
22:38like, "Oh, yeah, like, you know, we've
22:40got we might not have a residency
22:42opening this year. It might be next
22:44year, but let's, you know, send in your
22:46resume. let's let's get you um kind of
22:49coming back shadowing things like that.
22:52So um and that year they did not have a
22:54residency opening but I did go shadow
22:57with them like a couple of times a month
22:59>> and I'm like okay I really do like this.
23:01So the next year there was a residency
23:03opening. I applied and I I was accepted
23:06and um yeah I mean that's that's how it
23:09happened. But I think it was really the
23:11the background with my mom as a medtech
23:14because when I walked into the lab it
23:16felt like home, you know.
23:18>> Yeah. Yeah.
23:19>> And um and also just like I feel like
23:22it's so fun because we're like a
23:25consultant like you know the lab
23:28especially on like difficult cases where
23:30like the lab data is you know maybe
23:34there's things that are not quite so
23:37straightforward. um cytology findings
23:40that aren't straightforward. Maybe, you
23:42know, it's hard to incorporate the
23:44clinical context and everything
23:46together. It was last week that I had um
23:50oncology had a really difficult case and
23:53like they came down and like we all had
23:55a great conversation and the more we
23:58talked about it, the more I was like, I
24:00think we need to be worried about um a
24:02chronic myoid leukemia in this dog. like
24:04it was a crazy case uh chronic milo myoc
24:07leukemia and so um you know they were
24:11like yeah we were planning on getting
24:13aspirates of like the liver and spleen
24:15like looking for for cancer that's like
24:17infiltrative I'm like you know let's uh
24:19I'm like would you be willing to get a
24:20bone marrow as well and they were like
24:22yeah we were thinking about that I'm
24:23like let's go ahead and do that and then
24:25we got the bone marrow sample and that
24:26ended up being the the diagnosis so
24:30>> it's just really cool you know it's like
24:32taking complex cases and like just
24:35taking all the pieces and getting
24:36everybody in the same room and and
24:38having the conversation is so fun.
24:40>> Yeah.
24:41>> Well, you know, we have a lot of
24:42students that listen to the podcast and
24:45you know, when they hear words like
24:48clinical pathology, pathology,
24:50hisystopathology, cytologology, like how
24:52does it all fit together? Like when you
24:54go do residency, are you a pathologist
24:57and then you be, you know, like how does
24:59all of it work? you know, how can we
25:02help them understand kind of like all of
25:04that?
25:04>> This is such a great question. So, um,
25:08in, uh, in pathology, there are two
25:11divisions. Um, at least in veterinary
25:13pathology, there's essentially two
25:15divisions. There's clinical pathology
25:17and there's anatomic pathology. And so,
25:20um, anatomic pathology is, um, they
25:25they're primary for necropsies and for
25:28biopsies.
25:30Um so when you when you get a biopsy and
25:34put it in formalin and send that to a
25:36pathologist that's an anatomic
25:37pathologist. Um clinical pathology is um
25:43it's cytology it's the lab work uh you
25:47know so like CBC chemistry it's blood
25:49smears um the cytologies are a big thing
25:53and uh and also like advanced diagnostic
25:55testing like you know if you do
25:57endocrine if you do flowcytometry
26:01um PAR PCR for antigen receptor
26:04rearrangement serum protein
26:05electrofaresis like all those kinds of
26:07things um come under clinical pathology.
26:10So, um yeah, it's um you know, that's
26:14kind of the distinction. Anatomic is is
26:16the necropsies. They do a lot of
26:18biopsies. They do special staining
26:20things like that on the on the biopsy
26:22samples. And then clin is a lot of the
26:24the fluids, the blood, the body cavity
26:28eusions, the cytology samples, needle
26:30aspirates, impression smears, things
26:32like that. So, yeah. And and you do have
26:34to pick one, you know, like there are
26:37people that are double boarded. Um
26:41doing two residencies. Yeah. Wow. Um but
26:44generally speaking, you pick one, you do
26:45one at a time. So um so clinical
26:48pathology, you're going to go into the
26:49clinical pathology residency. If you
26:51want to do anatomic, you go into the
26:52anatomic pathology residency. And the
26:55way it works is everyone takes there's
26:59two parts of our boards. Everyone takes
27:02phase one together. That is the general
27:05pathology um phase one boards and then
27:08the second phase is your specialty. So
27:11it's either clinical pathology or
27:12anatomic pathology for phase two.
27:15>> Okay.
27:16>> Gotcha. Okay.
27:18>> That's good. That makes sense.
27:19>> So, uh kind of bringing it to like our
27:22everyday life, we Dr. King and I are
27:25general practitioners. So there's
27:27countless times where we'll submit a
27:29sample to have it analyzed. So like what
27:32information from a submitting
27:33veterinarian can um sorry where am I at?
27:37Sorry. Yeah. What information from a
27:39submitting veterinarian most improves
27:40the accuracy and usefulness of a
27:42pathologist interpretation? So what is
27:44best from us to give to you to then
27:46hopefully lead to the best diagnosis?
27:49>> Okay, this is also a great question. Um,
27:52so the history and like what the
27:56information you give us, it can be so
27:59critical. Um, and I like to tell people
28:03because I've had people say, I don't
28:06want to give you too much information
28:08because then you're like not blinded,
28:10you know? Like I've actually had people
28:11say that and I'm like no no no we blind
28:15ourselves [laughter]
28:17like when I'm about to look at a sample
28:19like because I don't you know I don't
28:21want something to bias me. Mhm.
28:24>> So I will blind myself like I will look
28:28at like okay this is a sample from a
28:31liver or from a math or whatever and
28:34then I'll I'll turn the paper over and
28:37I'll look at the sample and see what I
28:39see kind of you know get an idea from
28:42that and then and then I look at the
28:45rest of the clinical history and the
28:46context and see if it makes sense you
28:49know and like most of the time it does
28:50sometimes it doesn't and then you're
28:52like oh what happened
28:53Um, but so the information that a
28:58practitioner can give us really helps us
29:01because what are what are we we're not
29:04just reading what's on the slide, right?
29:07Like if we were just telling you, oh,
29:09there's spindle cells here, like do with
29:13that what you want, like then we're not
29:15very helpful. Um but if if you tell us
29:18like this is like the the critical
29:22things would be like what is the sample
29:26you know what is going on with the
29:28patient. So, for instance, really
29:32helpful history if you had like a mass
29:34on a dog like a, you know, subcutaneous
29:36mass or whatever. Um, description of the
29:39mass,
29:41uh, as thorough as you can be, size, is
29:44it attached to underlying tissues, is it
29:46firm, does it feel soft, you know, like
29:48all those all those nice descriptors.
29:51Um, and then, you know, like what's
29:54going on with the animal? Does it seem
29:56to be bothering the animal or was it
29:58just instantly noticed? Did the owner
30:00bring the animal in for this problem or
30:04is this just instantly knowed on an
30:06exam? Um, are there any other diagnostic
30:11tests that could, you know, could help
30:13with this? So like sometimes if uh if I
30:17see a mass and you know maybe it looks
30:20kind of like a like a saroma or
30:22something like that and I say you know I
30:25have concerns uh because we can't
30:27cytology can't always tell like what
30:30type of saroma right we can say this
30:32looks ugly it looks like a saroma I
30:34can't tell what type but I might ask
30:36like did you take x-rays and does it
30:38look like there's bony involvement or
30:42you know have you have you done any sort
30:44of like um additional diagnostic test
30:46that might be able to inform the
30:48diagnosis. Um so just like a thorough
30:53history is helpful. Um sometimes I say
30:57more is better, but that's with a
30:59caveat. Um because [laughter]
31:01there's sometimes a trend where people
31:03just like copy paste like everything
31:04that's been done ever and I'm like, "Oh
31:07man, I can't like this is like three
31:09pages of [laughter] history." like and
31:11you gota like go search for like the
31:13relevant stuff in there. But but you
31:15know I always say like you know maybe
31:18three to four sentences concise but but
31:22relevant information that if you were
31:25the pathologist about to look at this
31:27sample and you had no idea what was
31:29going on with this animal, what would
31:31you want to know?
31:33>> Like that's what I tell people.
31:35>> So
31:36>> yeah.
31:36>> Yeah.
31:37>> So more than just
31:38>> mass on toe. You want to know that?
31:42>> Yeah. Yeah. And
31:43>> have you done that,
31:43>> Doug? Well, [laughter]
31:45>> I think I've done that before, but
31:47that's just because I'm in the running
31:50to the next room and I
31:51>> Yeah. The technician throws the the the
31:54submission paper in front of you. It's
31:56like fill this out and you're like,
31:56"Mass onto showed up yesterday."
31:59[laughter]
32:00>> Yes. And like
32:03>> I've been there. I've been there. I
32:05totally get it. But like it's just kind
32:07of like help me help you, you know? Yes,
32:10of course it makes sense.
32:11>> Yeah. If this is a like for instance, I
32:14mean we can all understand this one. If
32:16this is a mass on a toe that Oh, sorry
32:19guys. That was
32:22[screaming]
32:26>> I thought that was King yelling at us
32:28talking about mass on toe. I thought she
32:31was like,
32:33>> "Oh my god, no was sitting in the window
32:36and he was not having it. Somebody was
32:37running by and he was like,
32:39>> "That is really funny." Oh my god.
32:40>> So loud and obnoxious. Oh my god.
32:43[laughter] Okay. Sorry about that, Mr.
32:44Town.
32:44>> That's okay. That's okay. So, so we can
32:47all understand this one. If you had mass
32:48on toe and it's, you know, less than a
32:52centimeter diameter and has been there
32:54for a year and a half versus mass on
32:58toe,
32:59owner first noticed a week ago, has
33:02gotten significantly larger since then.
33:04like gives us a lot of information and
33:06context. Um I remember one time I had uh
33:10I had an it was a sample that said
33:12memory mass.
33:14>> That's all it said. And um I was like
33:17okay intact female memory mass kind of
33:20older dog. I'm like uh oh. And then I'm
33:22looking at I'm like you know like memory
33:25masses can be really challenging
33:26psychologically because there's a lot
33:28going on in the memorary gland. But um
33:30but it looked like epithelial cells.
33:33They didn't look crazy. I was like, uh,
33:35you know, epithelial neoplasm of some
33:37sort. And then I'm like, this animal's
33:39in the hospital. Let me just go look at
33:40it. And so, like, I got up and I went
33:43over to the the ward that this animal
33:45was in. I was like, "Hey, this one said
33:47memory mass. I'd just like to get an
33:49idea like what what are we talking about
33:51here?" And they were like, "Oh, yeah,
33:52yeah, yeah. Come on." And so they like
33:54get the dog to roll over and then
33:56they're like, "Hold on, hold on, hold
33:57on. I can find it. I can find it." And
34:00then they like feel around. And then
34:01they finally find it. It's like this big
34:04and it's like hiding somewhere. I was
34:06like, "Okay, all right. Thank you."
34:07[laughter]
34:09>> Like, it's probably an adenoma,
34:11>> right?
34:12>> But I'm like, I don't know if we're mean
34:14like gigantic ulcerated gross thing
34:16versus like, you know, little tiny
34:19nodule thing. So, anyway, the more
34:22information you can give to help paint
34:23the clinical picture, the better.
34:25>> Yeah.
34:26>> Okay, I get it.
34:28>> Very helpful. Absolutely. One of the
34:29things that you had mentioned earlier
34:31was your passion for teaching students
34:33and you know helping them learn the ways
34:36of clinical pathology. How do you what
34:38are some of the most effective ways that
34:40you found to teach students to to kind
34:42of grasp this information so that they
34:44develop confidence in in their
34:46interpretation of the information and
34:48also the recognition of patterns and you
34:51know all of those things so that they
34:53don't get overwhelmed or you know
34:55sometimes we all do this where it's like
34:58>> um you know this is what I see and so
35:00that's what it's got to be you know like
35:01they come to a conclusion very quickly.
35:03>> Um so h how do you how do you handle
35:05that? Um so for background I teach um
35:10the second year that students in their
35:13clinical pathology course uh we cover
35:15lab work um blood smears
35:19uh cytology um so you know that's it's
35:23kind of like the overview course and
35:24like um teaching them about how to
35:26interpret blood work and we do a lab
35:29with um microscopy as well. Um, and then
35:32I also see them, again, I see them a
35:35little bit in third year for like some
35:36cases, you know, case discussions,
35:38things like that. And then I see them in
35:40fourth year, um, when they're on their
35:41clinical rotations on our diagnostic
35:43pathology rotation. And so I get to see
35:46the progression. And I will say over the
35:48years, because I've been at Auburn now
35:50for eight years, and they do at Auburn,
35:52our residents are very involved in the
35:54teaching program. Um, so over time like
35:58I've essentially been teaching for eight
36:00years and even though I've only been a
36:02professor for two. Um, but I have I mean
36:07my teaching has evolved quite a bit
36:10since the beginning. But what I've found
36:13that really helps first off just from a
36:15general teaching standpoint is
36:19first off being an engaging lecturer
36:21like you know I I don't know how best to
36:25say it but it's like you know actively
36:28engaging the audience being a safe space
36:32for questions
36:34>> is so important
36:36>> because like I've been there you know
36:38and I've been there a lot of times like
36:40in a lot of different aspects. If you're
36:42afraid of being wrong, then that's going
36:46to limit you from learning as much as
36:49you can.
36:51>> So the my classroom is a place for any
36:55question and people know this like you
36:58know I feel like it's now well
36:59established. I'm like please ask the
37:01question because what I tell them is if
37:03you have the question chances are
37:04multiple other people in the classroom
37:06also have the question. So please do not
37:08feel embarrassed about asking a
37:10question. And I have gotten some
37:12fantastic questions out of this. Um, our
37:15our lecture course it's 130 students and
37:19they just I want them to feel
37:21comfortable asking even though it's in
37:24such a big you know it's in such a big
37:26lecture hall but
37:28>> that it's been so good um ever since
37:31like now I open up my first lecture and
37:34I say this is a safe space to ask
37:37questions. Please ask them. And I I
37:42don't you know, you never want to make
37:44anybody feel
37:46dumb. Like you never want to make
37:48anybody feel like that. And they are not
37:50I mean like these are very intelligent
37:51people. But there, you know, sometimes
37:53people are afraid of asking because
37:54they're like, "Is this a dumb question?"
37:56I'm like, "There are literally no dumb
37:57questions. You've never seen this
37:58material before. Like please like ask."
38:02You know? So, um the other thing I like
38:04to do is give interactive
38:07um kind of exercises, like just short
38:09exercises during class. So, like let's
38:12say we cover a big topic and it's a lot
38:14to digest and I'll say we're going to
38:17take the next
38:1960 seconds and I want you to write down
38:21the key points that you remember from
38:25what we just said. Like, what do you
38:27think is the most important stuff that
38:29we just said? Um,
38:31>> and then sometimes that lets them think
38:34like, ooh, this was important. Wait, I
38:36don't actually understand that fully.
38:38And then we get some questions out of
38:39that.
38:40>> Or sometimes it'll be like a little
38:42interactive like, okay, we're going to
38:43take 60 seconds. We just talked about
38:46this really kind of dense concept. I
38:48want you to take 60 seconds and I want
38:50you to turn to your neighbor and explain
38:52it to them.
38:54>> And then they're like, they they turn to
38:57their neighbor, they [laughter] start
38:58trying to explain it. Some of them, you
38:59know, explain it and they they feel like
39:01they get it, but sometimes a question
39:03comes out of that. Like the neighbor is
39:04like, "Wait a second. I don't really
39:06understand that." And they're like,
39:06"Hold on, I don't really either. Let's
39:08ask the question." So,
39:10>> um, you know, things like that during
39:12class, I just that's
39:16so fun. Like, it's so fun when you get
39:18like the interactive discussion. Um, and
39:21I feel like second year we're just it's
39:24like a fire hose of like everything you
39:27need to know about lab work.
39:29>> And then third year we've got an
39:32elective course where where we go over
39:34chemistry cases and that allows them to
39:37kind of like put everything together.
39:39And then senior year is where everything
39:41I mean it like really comes together.
39:43So, you know, especially depending on
39:45where they get diagnostic pathology in
39:47their rotation schedule. It might be the
39:50first rotation, it might be a middle
39:52rotation, it might be the last rotation.
39:54Um, but it's, you know, it's just
39:58interesting to see even how they evolve
40:00over the year, like between different
40:03rotations. Like the first year they're
40:04like the first the first rotation
40:06they're like, "Oh man, this class was
40:08two years ago and I don't remember
40:09anything." And then the middle
40:11rotations, they're like, "Oh, yeah. I
40:13had internal medicine and I had
40:14emergency and critical care." Like, you
40:16know, we discuss some of this stuff, but
40:18then we still pick up on things that
40:20like, you know, they're just gaps that
40:23>> we can't cover everything all the time.
40:25So, like, you identify the gaps and you
40:27and you try to clear up those
40:28confusions. So, um, yeah, it's I love
40:34teaching like I love it so much. Um, but
40:38making sure coming back to the question,
40:40making sure that they don't jump to one
40:42thing too quickly. I think what what I
40:46like to do is when they go over cases,
40:50I I tell them how I do it and I'm like I
40:54look at I kind of break it down into the
40:56system. So, let's say it's like a
40:57chemistry case or like a CBC in
40:59chemistry. I break it down into like the
41:02systems and I'm like okay on a chemistry
41:05you have proteins, you have renal, you
41:07have enzymes, you have minerals, acid
41:11base, energy metabolism and
41:13electrolytes. Like you've got all these
41:15systems and I'm like go through and like
41:17kind of draw it out for yourself like
41:20what systems have something wrong with
41:22them. You know what's out of the
41:24reference interval
41:26>> somewhere? Write those down and then try
41:29to decide like always bring it back in
41:32context to the patient. You know like in
41:35context to the patient which of these
41:36could be the primary problem and which
41:39of these are like secondary issues that
41:41develop from the primary problem. So you
41:43know if you had a kidney failure you
41:46could expect electrolyte problems and
41:48you could expect acid base problems. So,
41:51you know, it's like there's primary
41:52kidney failure and then secondary would
41:54be your electrolytes and acid base.
41:57>> Um, or there potentially two primary
41:59problems, you know, something like that.
42:01And I I just try to get them to like see
42:03it holistically.
42:05>> Um,
42:06>> yeah. In terms of cytology, I always
42:09tell them, don't go too close too quick.
42:11Like stay stay at your low power
42:14objective 10x, you know, get the overall
42:18well. We talk about we joke about vibes
42:20in in cytology a lot, but get the
42:22overall vibe and then
42:25>> and then go closer as you need to. Don't
42:28like everybody always wants to go like
42:29see all the cellular details really fast
42:31and you could so miss something so
42:33easily that way. So yeah.
42:36>> Okay.
42:37>> So I mean a lot of that and I remember
42:40this from school as well that holistic
42:42approach right? Yeah. You may in one
42:43class be learning a specific like just
42:46about the CBC that day, but to your
42:48point, okay, now you have to as you
42:50learn and go through school and get in
42:51practice, what does that mean with
42:53what's attached or how it's attached to
42:55this animal?
42:57>> And I think getting into kind of the
42:58bigger discussion or a goal of this
43:00discussion we had of AI, like
43:03>> AI is great for certain things, right?
43:05And I think that was a in reference to
43:08cytology, clinical pathology, that was
43:10the main thing we were going to talk
43:11about today. But I also find as you're
43:13describing these things and how the
43:16example you had of going in the hospital
43:17to go see that patient and then apply
43:20the diagnostic approach or report is so
43:23crucial to what we do in our field
43:25because our patients cannot talk to us.
43:28Like that is a huge thing. We have to
43:30use all of the elemental pieces not just
43:34one portion of it. And I think with AI
43:37though it's helpful in some ways. like
43:39sure I like using it for um my notes
43:42because there's not much other thing
43:44they just have to spit it just has to
43:46spit out what I'm telling it to.
43:47>> They have to think that critically.
43:49Whereas in situations in your field like
43:52your specific field um of clin it's like
43:56>> no there's more to it. It's not just
43:58like punch in some things and let's see
44:01what spits out. So kind of going into
44:03that um kind of question real there like
44:08I guess starting off with the basic like
44:11people say like what when people say
44:13that AI can read a slide like what is
44:17technology actually doing in that
44:19scenario like how can you relate it to
44:20what you're doing
44:22right um so when they say that AI is
44:27quote unquote reading a slide we're
44:30talking about image recognition
44:32Um and machine learning in terms of it's
44:35an aspect of you know like image
44:38recognition is an aspect of the machine
44:40learning and so what it's doing is it's
44:44taking an image so like let's say it's a
44:47scan cytology slide it's taking that
44:51>> interpreting it as an image in terms of
44:53pixels and then then it sees the pixels
44:56and then neural networks are used to
44:59determine patterns.
45:01in the pixels. So, it might be looking
45:03for like a shape or a pattern or an edge
45:06or something like that. And then it's
45:08taking that information and matching it
45:12to what it's been trained on. So, it's
45:17looking for something that it's
45:20previously been trained on that matches
45:23that image. Mhm.
45:25>> And so, you know, like that's that's
45:30well and good, but it also it can't do
45:34human judgment.
45:37So, that's I think that's the limitation
45:40in AI. And when we start talking about
45:43diagnostic testing,
45:46>> um image recognition, it does have a lot
45:48of potential, you know, but but when we
45:51start creeping into the world of like
45:53human judgment, that's when um that's
45:56when we we need to start asking
45:57questions, I think. But to answer the
46:00question, that's how it's reading a
46:01slide is essentially taking the pixels
46:04um using the neural network to create
46:06patterns and then matching it back to,
46:09you know, an image that's been trained
46:11on. Sure.
46:13>> So, is it fair to say Oh, go ahead,
46:15King.
46:15>> Go ahead, Doug. Go. Go ahead.
46:17>> I was just going to say, is it fair to
46:18say like as you scan a slide, you're
46:20making the judgments as you're scanning
46:21it, or would you say that you kind of
46:23post-process after looking at
46:25everything?
46:26>> It depends. Um, so as with many things
46:29in veterary medicine, it depends. Um,
46:32>> so, so if it's like I mean, we
46:34definitely have those diagnoses that we
46:36we joke that they're brain stem
46:38diagnoses because it's like you don't
46:40even have to think about it. it just
46:41goes to the brain stem and then back. Um
46:43you know like mass cell tumor or um like
46:47large cell lymphoma or you know some
46:50ugly cancer like you know these are kind
46:53of like brain stem things a melanoma
46:55with a ton of black pigment in it. Okay
46:57like that's a melanoma
46:58>> but sometimes you get those samples and
47:01this is this is more often like I would
47:03say the brain stem diagnoses happen. We
47:05love it when that happens you know we're
47:07like okay cool I know this. Um but then
47:10there's a lot that take a lot of a lot
47:14of um additional judgment, a lot of
47:18incorporating all the information. And
47:20so, you know, a lot of times I I might
47:23see a sample and it's like, okay,
47:26I'm seeing
47:28um for example, one of the challenging
47:30diagnoses is is sometimes um like
47:33granulomitus or pyranolomitus
47:35inflammation
47:36>> because macrofasages
47:38like we have a joke in clin that you'll
47:41spend three years trying to figure out
47:42what a macroofage is and isn't. Um like
47:46it's like they can look all kinds of
47:47different ways. And so in pyoganolomitus
47:49or granulomitus inflammation
47:51macrofasages can become bi-ucleiated
47:55multi-ucleiated they can they can like
47:58form these like sensitia that can look
48:01like epithelial cells they can look like
48:03meal cells they can look like all kinds
48:05of stuff and it's one of those things
48:07where as you're looking at I remember
48:09one it was an anal gland mass this was
48:12like I was a secondy year resident or
48:14something it was an anal gland mass and
48:17I'm looking at it and it was like it it
48:21what it ended up being was
48:22pyioanolominous inflammation. But the
48:24first thing I thought was like oh this
48:26has to be neoplastic because it's like
48:28multi-ucleane cells all kinds of like
48:31crazy stuff going on. And then the more
48:33I looked at I'm like wait there's a lot
48:35of neutrfils in here. This is
48:36inflammatory in some way. Okay you can
48:38have that like inflammation with a
48:40tumor. Like that's not uncommon. But
48:43then the more I looked at it, the more I
48:45realized like all of the nuclei looked
48:47fairly similar and it they looked like
48:51macrophagages. And every now and then
48:52I'd see one that was like very
48:54definitively a macroofage. Like it's
48:55vaculated, it's eating things like
48:57that's a macroofage. And then I'm
48:59comparing that nucleus over to the other
49:01nuclei in these crazy multi-ucleated
49:03cells and I'm like dang these are all
49:05macroofagages and this is pyiograinous
49:08inflammation. And then I found the fungi
49:10after that. So, you know, it's like,
49:12"Oh, now I find the little organisms."
49:14Um,
49:16>> yeah. So, you know, that's just like
49:18>> sometimes that's how the like cytology
49:20case builds. You like some of them you
49:23can look at in 10 seconds and know what
49:25it is and some of them, you know, you
49:28have to cut yourself off at some point.
49:30You like stare at it for like 10, 15
49:32minutes and then you're like,
49:34>> "Okay, all right. it's time to like sit
49:36back, put this information together, and
49:39give them an answer. So,
49:42>> yeah.
49:42>> Yeah. You know, I've heard some of the
49:44pros or like when people are talking
49:46about using AI for cytology or for
49:49reading cells, like some of the
49:51defenses, if you will, or pros that they
49:53throw out there, like, you know, the
49:55machine doesn't get tired, you know,
49:57like from 8:00 a.m. to 5:00 p.m. the
50:00same you know like whereas human having
50:03human eyes doing something for so many
50:06hours a day do they is there something
50:08that occurs like from the beginning of
50:09the day to the end of the day where then
50:11the diagnosis isn't as accurate at the
50:13end of the day as the beginning of the
50:14day or you know they look at so many the
50:18the AI looks at so many more cells then
50:21because it can do it in such like I
50:22guess like it's doing it like a however
50:24it's reading the thing you know um and
50:27so I'm like you know I could see why
50:29that would be appealing to people like
50:31you know in terms of because they're
50:33selling it as a more accurate way of
50:37getting an interpretation or a
50:39diagnosis. Um how do you what is your
50:43feeling about that as a as a clinical
50:45pathologist? Like what how do you feel
50:47about those statements when they're
50:49making those types of statements?
50:51Um so
50:54when they say so my first question when
50:58I when I hear statements like that is I
51:03ask just you know myself really like I
51:06ask anybody who listen um
51:10how is the machine being trained
51:14>> and how is it being validated?
51:18So,
51:19and those are questions that we don't
51:21know right now.
51:23>> Um, because there have been no
51:25publications
51:27of the methodology behind the training
51:29or the validation on any machine at this
51:32point. And I'm just talking about AI
51:34cytology in general. Like there have
51:36been no no publications, no white paper
51:40publications validating any of the
51:42machines or describing the training
51:44methodology. So, if we're gonna say it's
51:47more accurate than a pathologist,
51:51first off, we have to validate it that
51:53way.
51:54>> We have to train it that way. We have to
51:55validate it that way. And you can do
51:58that. I mean like in theory
52:01>> um
52:04>> you I just I know the nuances of
52:07cytology
52:08>> and I just imagine the amount of work it
52:12would take because like I just break it
52:15you know every tumor type
52:18>> it's matching it back to what it's been
52:20trained on
52:21>> trained. Yeah.
52:22>> And so if we ask what has it been
52:24trained on are we talking about images
52:26in textbooks? Are we talking about real
52:29world images? Are we talk like what are
52:31we talking about exactly? And if we're
52:34gonna say it has it is more accurate
52:37than a pathologist, how did we validate
52:39that? Because you would have to if
52:41you're going to say it's more accurate
52:43than a pathologist, but you've only
52:45trained it using clinical pathologists.
52:48It cannot be more accurate.
52:50>> More accurate than Yeah.
52:51>> It can't be more accurate. like and
52:53>> right
52:54>> you would have to essentially train and
52:56validate it based on the ultimate
52:59diagnosis.
53:01So not like what the pathologist said. I
53:04mean you'd want to also know that but
53:06you would have to train it on what the
53:07ultimate diagnosis eventually is. So
53:09that would be like necropsy or hystopath
53:11[clears throat]
53:12or some sort of definitive like we got
53:14an answer on what this is.
53:18>> That's the only way you're going to be
53:19able to say it's more accurate than a
53:21pathologist. Yeah.
53:22>> Um,
53:24>> so I I just I ask a lot of questions
53:28when I hear
53:31>> these statements, you know, it's the
53:33training, the validation. We need to
53:35know these things and we don't know
53:37these things.
53:39>> Um, the other thing, so when I, you
53:42know, I started getting really
53:43interested in this a few months ago
53:45because it was kind of like I was
53:47hearing things from general
53:49practitioners. They're my friends. I was
53:51being sent slides that were read as
53:53inconclusive
53:55>> by AI and they were asking, you know,
53:57they were just like, "Hey, I've got some
53:59pictures here. Can you look at this?"
54:02>> And the inconclusive ones were actually
54:04very diagnostic. Like I had a um one was
54:08the the question was lymphoma or
54:10reactive lymph node like enlarged lymph
54:11node is it lymphoma? Is it reactive?
54:14very obvious reactive lymph node like
54:16like I said brain stem diagnosis like
54:18this you know one of them was a melanoma
54:22that was very pigmented like very
54:25obvious oral mass very pigmented very
54:28obvious melanoma one of them was a mass
54:30cell tumor that was not well granulated
54:33>> so like brain stem diagnosis for any any
54:36clinical pathologist but like AI's
54:39reading is inconclusive and so that just
54:41makes me ask like okay well how's it
54:42been trained
54:44you know, and what is it validated for?
54:46A lot of a lot of, you know, current AI
54:48cytology tech is only validated for
54:52certain types of tumors.
54:56And I mean, at present, it's I don't
54:59know how much it's adding because it's
55:00like it's validated for lymphoma and
55:02massel tumor. A lot of people can tell
55:04that without a clinical even. So, you
55:07know, so it's one of those questions.
55:08It's like, okay, that's great. But um
55:11but I just I I think the question we
55:13need to ask is how is it being trained?
55:15How is it being validated? Because
55:16there's so many subtleties to clinical
55:18pathology. So many different tumor
55:20types, so many different types of ways
55:22those tumors can look. If I just take a
55:24mass cell tumor, like you think about
55:26the well- granulated ones that like
55:28everybody knows, like
55:30>> we can give that to to a second year vet
55:32student and they're like, "That's a mass
55:33cell tumor." You're like, "Cool. Yeah,
55:35it is." like the well- granulated ones.
55:37Super simple. But how many different
55:38ways can mass cell tumors look? Like we
55:42have the nice well granulated ones. We
55:44have the poorly granulated ones.
55:47Sometimes it's hard to tell like you're
55:49like this is round cell and I think it's
55:51a mass cell tumor but um you know it it
55:54looks
55:55>> not normal. Um sometimes I have seen
55:58mass cell tumors that are because mass
56:01cell tumors um very commonly cause
56:03fibroplasia and cause a lot of tissue
56:06damage and you'll get fibroplasia in the
56:08mass cell tumor. And so I've gotten some
56:10mass cell tumors that were almost all
56:12spindle cells and then just like weird
56:15>> little groups of mass cells that you
56:17know you had to go hunt for. So they can
56:21and then and then you get into the
56:22staining diffquo
56:25which is what we use in a lot of clin
56:27labs like diffqu has this this uh fairly
56:32well-known issue where it doesn't stain
56:33mass cell granules all that well so like
56:36sometimes you don't see any granules at
56:38all
56:38>> and so like they can look all kinds of
56:40different ways and that's that's where I
56:43asked the question like how is it being
56:44trained is it being trained on a
56:45textbook because if it's being trained
56:47on a textbook it's not going to pick up
56:49that other stuff. If it's being trained
56:50on real world samples, then okay. Like,
56:53you know, then then we're talking. But
56:55it's got to
56:56>> it, like I said, going back to how how
56:58it recognizes things, it matches the
57:00pattern to what it's been trained on.
57:02>> And so, if it hasn't seen that
57:05particular
57:07iteration of whatever the tumor is, then
57:11then it may miss it. Um
57:15so
57:16>> so are you surprised are you surprised
57:18with some of the legal things that have
57:21been popping up in regards to um
57:24clinicians trusting a diagnosis from AI
57:28and then you know finding out that no
57:30that's in fact not what it was. Does
57:32that surprise you or are you like you
57:34know
57:36>> I'm not surprised. Um,
57:39I I um this is this is probably a place
57:43to say my opinions are my own and do not
57:46reflect those of any past or current
57:48employer. Um [laughter] I I'm not
57:51surprised at what's happening because
57:53okay
57:55I I want to start in the right place for
57:59this. Um um the first thing I want to
58:02say I this is something that I like I
58:05said I started looking into this you
58:07know a few months ago and this is
58:09something I didn't know until then and I
58:12think it's important for veterinarians
58:14to know just in a general sense the FDA
58:18does not require any sort of validation
58:22for animal medical equipment.
58:25Oh, interesting. Okay. I don't know if I
58:28knew. I didn't know that.
58:29>> I know. No, like nobody knows that.
58:31That's why I'm saying it now because I
58:33It was news to me and I'm like, if it
58:35was news to me, then it's news to a lot
58:37of people. So, I just want to read y'all
58:39the statement.
58:41>> This is on the FDA website. How FDA
58:44regulates animal medical devices. Okay.
58:48>> Pre-market approval is not required. The
58:51FDA does not require submission of a
58:53501k, PMA, or any pre-market approval
58:56for devices intended for animal use.
58:58Device manufacturers who exclusively
59:00manufacture or distribute animal devices
59:03are not required to register their
59:06establishments or list animal devices
59:09with the FDA and are exempt from
59:12postmarket reporting. It is the
59:14responsibility of the manufacturer
59:17and/or distributor of these articles to
59:19asssure that the animal devices are
59:21safe, effective, and properly labeled.
59:24The FDA encourages veterinarians and
59:27animal owners to report adverse
59:29experiences and product defects
59:32associated with animal devices.
59:35Huh.
59:36>> So what that's saying is that the FDA
59:39does not have any sort of regulatory
59:43requirement for validation or approval
59:46for animal medical devices which is what
59:48things fall under like I mean any lab
59:51work machine um you know AI anything uh
59:56syringes need you know like any sort of
59:58equipment you would use it doesn't it
1:00:01doesn't require any sort of
1:00:03>> pre-market approval validation
1:00:05and actually says the validation is on
1:00:08the manufacturer or distributor.
1:00:12>> Yeah, that's kind of interesting.
1:00:15>> That's fascinating.
1:00:17>> Yes.
1:00:20That's like
1:00:23I don't know what to say about that.
1:00:25[laughter]
1:00:27>> So, like
1:00:29how do we [laughter] I don't know. How
1:00:32do we know it's all good and dandy? I
1:00:34mean,
1:00:35>> you don't.
1:00:37>> You don't. That's right.
1:00:38>> You just shattered my world, Dr. Towns.
1:00:41I don't know what to
1:00:44Sorry. People need to know.
1:00:46>> A very suspicious person. She's gonna be
1:00:48like, I can't send anything this way,
1:00:50this way. [laughter]
1:00:52>> I've already got my thoughts and
1:00:53suspicions on when and where I should
1:00:55use AI and stuff in clinical pra from
1:00:58just like practicing alone when it's
1:01:00become available. There are times where
1:01:02I'm like this makes no sense or just to
1:01:05your point of a cytologology didn't was
1:01:07um what you say non-diagnostic or not a
1:01:10goodam sample. I then looked at it I was
1:01:12like there are cells here sent it out to
1:01:15a person and they came back with a
1:01:17report. So I'm like that's not true.
1:01:19>> So don't even get me started on your
1:01:21analysis.
1:01:23So [laughter]
1:01:26>> yeah,
1:01:26>> we'll keep that there. So
1:01:28>> fascinating.
1:01:30>> Yeah. Yeah. Um, I want to I want to take
1:01:34us back. So, you know, I've had people
1:01:36ask me, "Okay, are you anti- I am not
1:01:39anti-AII?" Like, let me tell you that.
1:01:41Like, I think AI can be used
1:01:44effectively. I do not think it should
1:01:47replace a specialist.
1:01:50>> Like, I let me tell you, okay, so I'm
1:01:53going to take us back in history a
1:01:54little bit to hematology machines. This
1:01:57is something that a lot of people don't
1:01:58know. Um, I know because my mom was a
1:02:01medtech
1:02:03>> when the first hematology machines were
1:02:06coming out. So before before he
1:02:09machines, we're talking about like our
1:02:11little benchtop CBC analyzers,
1:02:14>> you know, um, and then diagnostic labs
1:02:16have like big real fancy um, hematology
1:02:20analyzers. But but let's take it back to
1:02:22the 1960s
1:02:24when um before the advent of a machine
1:02:28analyzer. Um cells were handcounted.
1:02:32>> Right.
1:02:33>> Right. So that's obviously, you know,
1:02:35it's got to have a lot of um a lot of
1:02:37variation, a lot of inaccuracy, but it
1:02:39was the best thing we had at the time.
1:02:42>> Then there was the advent of the first
1:02:43machine, which was the culture counter.
1:02:46Um, and so it was the first way that we
1:02:48had to like automate account.
1:02:51Um, and I have actually talked to people
1:02:53from this this era when the culter
1:02:56counter came out and they were like at
1:02:57first it was terrible. Like you know our
1:03:00our counts were way more accurate. Like
1:03:03the culter counter was getting all kinds
1:03:04of things wrong. But it went and lived
1:03:07in diagnostic labs with med techs and
1:03:10pathologists first. Like that's where it
1:03:12went. it lived there and they, you know,
1:03:15they gave feedback, you know, hey, you
1:03:19know, we're getting really inaccurate
1:03:20counts. Okay, back to the drawing board,
1:03:22you know, and then so, so after that,
1:03:25that was the 1960s, 60s to 70s. Um,
1:03:28after that, you know, the machine gets
1:03:31better. And so over the next decades,
1:03:34the machine gets better, the machine
1:03:35gets better, and it and it evolves into,
1:03:38you know, um, we're getting to the 1990s
1:03:41now. Now we have a machine that um it
1:03:44has lived in a diagnostic lab with
1:03:46pathologists and medte for this amount
1:03:48of time and now it's evolved into
1:03:49something relatively close to what we
1:03:51know today. Um and so you know it's
1:03:55accurate people trust it like and and
1:03:58even with that we still do blood smears
1:04:01on every single CBC that comes through a
1:04:04lab. So, someone will look at a blood
1:04:07smear on every single CBC that is run by
1:04:10a lab, a di a big diagnostic lab. Um,
1:04:13and that's once again just having human
1:04:16validation. We catch mistakes all the
1:04:17time, right? Like we still catch
1:04:19mistakes or we still catch things it
1:04:21misses, the things that we know the
1:04:23machine isn't capable of.
1:04:25>> So, we know those are limitations and we
1:04:27go look for them on the slide.
1:04:30Um, so it took up until the 1990s
1:04:33before a company, any company, you know,
1:04:37there were several of them, felt
1:04:38confident enough in their product to put
1:04:41it in a in a a general practice setting.
1:04:45And this is in human medicine as well.
1:04:47So it lived in the diagnostic lab with
1:04:50med techs and pathologists until the
1:04:53point where the technology became
1:04:55accurate and consistent enough that it
1:04:57they deemed it ready for general
1:05:00practice settings and not to have like a
1:05:02pathologist or medtec saying they're
1:05:03reviewing everything. And so that's when
1:05:05it was 1990s. So that was 30 to you know
1:05:0930 years before we had machines in labs
1:05:14or in general practice uh settings. And
1:05:17then over like we still we still say hey
1:05:20look at a blood smear on every single
1:05:21one. I know y'all I know it's hard. I
1:05:23know it's hard personally. I know it's
1:05:25hard but um but yeah that's a discussion
1:05:28for another day.
1:05:29>> [laughter]
1:05:30>> Um, so and then then we um you know we
1:05:33just keep having advances and they get
1:05:35better and better and now even with as
1:05:38good as they've gotten like I mean we
1:05:40have some fantastic machines both in our
1:05:44diagnostic labs and in the general
1:05:46practice there's some really fantastic
1:05:48hematology analyzers out there but they
1:05:50still there are still things that we
1:05:52know they miss
1:05:53>> and there are still things that we have
1:05:55to look at a blood smear for and like
1:05:59it's using machine technology to count
1:06:02the cells. So if it gives you a lucasite
1:06:04differential, it's using flowcytometry
1:06:06to get that. And sometimes those cells
1:06:09for whatever reason overlap and so the
1:06:12way the flowcytometry reads it, it's
1:06:14incorrect. And so that's why we look at
1:06:16a a blood smear on every one and we get
1:06:18our own count. You know, we know it's
1:06:21not as accurate as the machine can be,
1:06:24but we also know that if the machine
1:06:26miscatategorized something, we can catch
1:06:28it that way.
1:06:30>> So, yeah, it's and so that was the
1:06:34timeline of the hematology analyzer and
1:06:36how it came about. So, I like to compare
1:06:39that to what's happening now is there's
1:06:44the tech.
1:06:46I feel like it has gone straight into
1:06:48the general practice
1:06:50>> without going through that step that the
1:06:53hematology analyzers did. And so,
1:06:56>> so the question is like like with the
1:06:58hematology analyzers, we know where the
1:07:01limitations are
1:07:02>> and we know where the weaknesses are.
1:07:04And so when we when we check behind it,
1:07:07we know what we're looking for. Whereas
1:07:09with the current AI, I mean this goes
1:07:12for anything AI like radiology,
1:07:14cytologology, anything, we don't know
1:07:17specifically where the weaknesses are.
1:07:19And so if we get a result,
1:07:24what does it mean? like you know it's
1:07:27like I and we haven't seen any like I
1:07:30have not seen any validation that states
1:07:33any sort of sensitivity or specificity
1:07:36numbers for any various types of you
1:07:39know lesions. So like skin cytologology,
1:07:42ear cycology, back like you would have
1:07:45to do a sensitivity and specificity
1:07:46calculation for every one of them.
1:07:48Bacteria does it see it or does it not?
1:07:51Like you know mass cell tumor did it see
1:07:53it or did it not? what is the
1:07:54sensitivity and specificity on each one
1:07:56of them and we just don't know. We have
1:07:58no idea.
1:07:59>> So, um you know that's that's the
1:08:03unknown
1:08:05and I think that's where we really need
1:08:07more development and more information
1:08:10and more um you know more transparency
1:08:14on on the methods and and the
1:08:18development of AI. Um, but currently it
1:08:21doesn't exist with a pathology. Like I
1:08:23mean it I I haven't seen one personally
1:08:26like in a diagnostic lab. I'm sure maybe
1:08:28it has happened. I just it's it's not
1:08:31transparent. We don't know.
1:08:33>> So we we just it's like we're in the
1:08:36stage of we don't know what we don't
1:08:38know, right? And and that's where we're
1:08:40at. Do you think at this stage there is
1:08:42a valid place for AI at all other than
1:08:45in my opin not note takingaking I guess
1:08:48like the clinical pathology setting. Do
1:08:50you feel like there is a place for AI at
1:08:52all or we still more time?
1:08:55>> So
1:08:56>> if as of right now I have a lot of
1:09:01concerns about it um for the reasons
1:09:04I've stated. But if we if we took an
1:09:08approach like here's in in Jordan Towns
1:09:12is ideal world um [laughter]
1:09:15which you know whether or not it happens
1:09:17I don't know but but um the I think the
1:09:21best place that it could be used like I
1:09:25think there is a legitimate use for it
1:09:27if I had a machine that took some of the
1:09:33really easy stuff off that plate.
1:09:36you know, like it could screen it like,
1:09:39hey, screen for the things we know it
1:09:42can see, you know, whatever those may be
1:09:44at this time. I I know I've heard it's,
1:09:46you know, validated in whatever way for
1:09:49lymphoma and mass cell tumor or and
1:09:51lipoma or something like that. So, if
1:09:53you could take those things off of my
1:09:55plate and leave me time for like the
1:09:57more complicated stuff, like I don't
1:09:59have a problem with that, you know? And
1:10:01I still like personally I would still
1:10:03want to put eyes on it. But if it's like
1:10:05the machine say it's a mass cell tumor
1:10:06and I put it down I'm like yeah mass
1:10:08cell tumor. You know that would take me
1:10:09like 3 seconds. And if if the machine
1:10:11could go ahead and generate that report
1:10:14>> and like I just kind of like briefly
1:10:16read over it. I'm like yeah looks good.
1:10:18Like that would be a a good use of it.
1:10:21Like I think it would be great there and
1:10:22it would speed things up. You know all
1:10:24the claims that we hear, oh it can speed
1:10:25things up.
1:10:26Like I don't know that I've ever heard
1:10:28like I have you know how many of the
1:10:30cytologies are emergencies
1:10:32>> right you know like yeah like it's I
1:10:36mean there they exist you know like the
1:10:38septic abdomen okay we want to know if
1:10:40it's septic right now but like a lot of
1:10:42things are just not emergencies
1:10:45>> and I understand wanting to have a fast
1:10:47turnaround time and I do think AI could
1:10:50aid us in having a fast turnaround time
1:10:52but I don't think it should be at the
1:10:54expense of getting inaccurate wrong
1:10:58information, right? Like
1:11:00>> even if if I try to go too fast, you
1:11:03know, it's like,
1:11:04>> okay, well, that doesn't help anybody.
1:11:06So, if we could clear off some of the
1:11:08easy things and leave the pathologist
1:11:10time to um, you know, to look at the
1:11:14more complex cases, I think that would
1:11:15be great use of it.
1:11:17>> I agree.
1:11:19>> Yeah. I could also see potentially in
1:11:21the future like I wonder if it couldn't
1:11:24help maybe there's already like a
1:11:26platform that's doing this but like you
1:11:28know when you first get out of school
1:11:31and you have some really whacking labs
1:11:34and just to keep you from forgetting
1:11:36potential differentials
1:11:38>> like you know like hey these things
1:11:40should be you know like on your list
1:11:42like just like not telling you what to
1:11:44do but saying don't forget you have
1:11:46these three things and these three
1:11:48things can show up with this thing. So,
1:11:50don't forget about this. Like I that I
1:11:52think would be like really helpful for
1:11:54the younger veterinarians who are like
1:11:57questioning, should I think about that?
1:11:59Okay, like maybe I should
1:12:00>> and maybe don't have access to a mentor
1:12:01immediately, you know, like that's a
1:12:04tough place to be in. Like I've been
1:12:05there, you know, and it's like if you
1:12:08feel like you you you know, you don't
1:12:11have the guidance and you don't want to
1:12:13miss something and you've got something
1:12:15that's like
1:12:16>> pretty complex. No, I've also said that
1:12:18because like I mean how many times am I
1:12:21like looking at a complex cytology
1:12:23sample and it's something weird and I'm
1:12:26like can that happen? Let me do a lit
1:12:28search really quick. And then like I do
1:12:30a lit search, but it's like if you know
1:12:32if I had a tool that had been validated
1:12:35for veterinary medicine like a a lip
1:12:38review tool or something like that and
1:12:40it's like
1:12:41>> um can you know how often does this
1:12:44tumor happen in dogs on the toe or
1:12:48whatever like and and it's like um you
1:12:51know I've got like my main differentials
1:12:53but then sometimes it might be like hey
1:12:55don't forget about this one kind of
1:12:57weird one that you know isn't common but
1:13:00does happen. Like I think that could be
1:13:02a great place
1:13:03>> for the use of AI like like as a a
1:13:06friendly companion that's like would you
1:13:09consider you know like
1:13:12>> I agree as like a friendly companion
1:13:14because I worry that if people rely on
1:13:15it so much just for the sake of moving
1:13:18faster and burn and churn if you will
1:13:22then you're going to lose that knowledge
1:13:23and part of your brain that's
1:13:25functioning for that specific issue.
1:13:27Like I feel like if as a general
1:13:30practitioner too, if I'm looking at
1:13:31something under a microscope and I'm
1:13:33uncomfortable either the way the dog or
1:13:36cat is acting or what I'm seeing I'm not
1:13:38like able to call it and do a certain
1:13:41treatment plan, I'm going to submit that
1:13:43and then get re affirmation like what
1:13:45you guys are saying like this should be
1:13:46on your list or hey no, I was thinking
1:13:48of it correctly. I'm going to put that
1:13:51now in my toolbox of experience I can
1:13:53use to help my patients. Whereas if
1:13:55we're just constantly just putting it in
1:13:57a machine saying, "Oh, I didn't even I
1:14:00didn't even know what those cells looked
1:14:01like. I just put it in the machine."
1:14:03Like I feel like that's such a
1:14:04disservice to all the animals we could
1:14:06help, especially if an owner cannot
1:14:08afford to do AI or something to that
1:14:10extent,
1:14:11>> right?
1:14:12>> So, it's
1:14:13>> I I I definitely I see what you're like
1:14:16what we're saying of like a helper
1:14:17friend, but we need to get further than
1:14:20where we are right now to achieve that
1:14:23essentially.
1:14:24Definitely. Yeah. And like, you know,
1:14:26I've I've seen um you know, I
1:14:30>> I just always urge caution with things
1:14:32because like sometimes I'll I'll put lab
1:14:34work into AI just to see what it comes
1:14:38up with or I'll put I'll put something
1:14:40that I know people make mistakes with
1:14:43and I'll see does the AI get it right or
1:14:46does and like so far the AI has not
1:14:49gotten it right in a lot of those cases.
1:14:51Like I I feel like right now where we're
1:14:53at with AI and lab work
1:14:56>> is it still
1:14:58>> it's it still sees things independently
1:15:01>> like it doesn't look at it all together.
1:15:04So it sees like amia. It's like oh you
1:15:06know kidney stuff or whatever but it's
1:15:09and then just it's kind of like step
1:15:11away. It's like telling you what's there
1:15:13but not telling you like the whole
1:15:15picture and it may not incorporate
1:15:18certain things in you know that that
1:15:21need to be incorporated in. So
1:15:24>> yeah and I think it could get there. I
1:15:26just you know it's like how do we
1:15:28validate? It comes back to that question
1:15:30like
1:15:30>> how are we training it? How are we
1:15:32validating it? And these are things we
1:15:34need to know like did we did we do our
1:15:37due diligence in validating it make sure
1:15:39it's saying the right things and then
1:15:41you know you get into the whole like
1:15:42hallucination thing like you know it's
1:15:45like how do we know when it's
1:15:46hallucinating I went to a fascinating
1:15:48webinar on um AI in research
1:15:53um like generating research papers and
1:15:56that was crazy because um it was a study
1:16:00on um just people that have used AI to
1:16:05um get their references section, like if
1:16:08they're writing a scientific manuscript
1:16:10and they need a references section and
1:16:11they've been like kind of, you know,
1:16:14doing searches using AI and it's giving
1:16:16them, you know, like, oh, this is a
1:16:18great paper, you should go look at this,
1:16:20you know, or whatever. So, then it asks
1:16:22then they asked at the end, can you
1:16:24generate a references section for me
1:16:26based on the papers we've talked about?
1:16:28And they found that in I'm not going to
1:16:33say a huge number of cases, but in a in
1:16:35a decent number and it's it's
1:16:37increasing. The more we see AI used, the
1:16:40more the the number increases, but AI is
1:16:42actually fabricating references.
1:16:46>> Yeah.
1:16:47>> Wow.
1:16:48>> Like making up names, titles, years.
1:16:54And so they did this retrospective study
1:16:56and went and found fabricated
1:16:59references in like a certain proportion
1:17:03of it wasn't a ton like I said but the
1:17:06AI had very clearly been fabricating
1:17:08references. Oh my god like just making
1:17:10them up and um and the it's like the
1:17:13question like the next question I ask
1:17:15just personally is why?
1:17:17>> Why? Yeah. Exactly.
1:17:18>> Do that.
1:17:19>> Yeah.
1:17:20>> Yeah. Well,
1:17:21>> and then like if it will do it for just
1:17:23random references, what will it do
1:17:27when you ask it a question? That's why
1:17:29like when I hear like the notes thing,
1:17:31I'm like, go read those.
1:17:34>> Oh yeah, we read them. Oh yeah.
1:17:35[laughter]
1:17:37>> It's like, oh yeah, just document what I
1:17:38say. Okay, that's great. And then it's
1:17:40like it completely puts something in
1:17:42there.
1:17:43>> It will like a simple example is like we
1:17:46will do an annual and we'll do the
1:17:47heartworm test. will send out to the lab
1:17:49heartworm tests and uh whatever blood
1:17:52work and in the note in the email it'll
1:17:53be like we submitted these and they came
1:17:55back negative and I'm like no like I
1:17:58[laughter]
1:17:59>> that's what happened
1:18:00>> that is yeah that's a standard policy in
1:18:02our practice have to edit it and
1:18:04>> you have to edit them yeah
1:18:06>> yeah put things in there that's like
1:18:07it's benign I'm like I didn't say it's
1:18:09benign don't do that
1:18:11>> don't put those words in my mouth
1:18:12>> yeah
1:18:14no and that's the thing and so then the
1:18:15next question is okay if we use AI How
1:18:18much time are we having to spend going
1:18:19back and rereading it and making sure it
1:18:21says the right things when we could have
1:18:23just written it ourselves? Now, I am one
1:18:25person that spends way too long on
1:18:27medical records. So, I would probably
1:18:28use AI and be faster if I just read
1:18:30through and saw what it said. But, um
1:18:34talking extra if I don't have to in a
1:18:36day. And so, sometimes I don't do the AI
1:18:39and the purpose I should of going
1:18:40through everything because I'm like, I'm
1:18:42just tired. I'm just going to click
1:18:44generate and go back through and fix it.
1:18:45Like I know that's a waste, but
1:18:47[laughter]
1:18:47I don't want to talk anymore, let alone
1:18:49[clears throat] do a computer.
1:18:52>> Yeah. No, my favorite is when I I was
1:18:55trying to like it was like a football
1:18:57game or something last year. I was
1:18:58trying to Google like what TV station
1:19:02and what time this game was at or
1:19:05whatever. I can't remember. But I I
1:19:07asked it and it comes up with that
1:19:08little AI, you know, verb. And this game
1:19:12was like in a few days and it came up
1:19:15like like the the 2026 Packers versus
1:19:20Seahawks game was held on at 700 p.m. on
1:19:25August 21st and the score was 17-3
1:19:30Packers win. And I'm like it has not
1:19:33happened.
1:19:34>> Oh my goodness. just made it up like
1:19:37like and then you want to like
1:19:39screenshot it and see if it gets it
1:19:40right because then like oh god what are
1:19:42we doing now but yeah
1:19:43>> oh boy
1:19:45so bizarre I'm like did you just make up
1:19:48a score
1:19:51it's getting there
1:19:52>> it is getting wild
1:19:54that's that's my point with the FDA I'm
1:19:56like it's kind of the wild west a little
1:19:58bit so like
1:20:00>> they're going to have to do something I
1:20:02mean something have to change just
1:20:03because there's no you know oversight
1:20:07and so I mean
1:20:10>> so okay we talked about a lot so let's
1:20:12kind of go back through some takeaways
1:20:14from this whole discussion right so if a
1:20:17veterary clinic is considering using AIC
1:20:22platform what questions should its team
1:20:24ask before trusting the technology with
1:20:27the patients which we've kind of alluded
1:20:29to throughout this conversation but like
1:20:31to make it like
1:20:34straight straightforward, I guess, more
1:20:35straightforward as we can be,
1:20:37>> right? And um you know, I think be
1:20:40willing to ask the tough questions. Um
1:20:43be willing to ask like how was this
1:20:45validated? What was it validated on? And
1:20:48like I feel like it's now that we've had
1:20:51this conversation, those questions are a
1:20:54little bit easier cuz it's like if you
1:20:56don't if you don't know like how it
1:20:58should be validated or you know then you
1:21:01might not know what the right answer to
1:21:03the question is or how it should be
1:21:05trained, how it should be validated. But
1:21:08um you know I would definitely ask those
1:21:10questions. If they say it's, you know,
1:21:13capable or it can do these things, I
1:21:15would ask for the data on it. Um,
1:21:19you know, if I would just ask for the
1:21:22data and and ask the questions like how
1:21:25was it trained? How was it validated?
1:21:26How do we know that we can be confident
1:21:29in these answers?
1:21:31>> And then I would also ask, do you stand
1:21:34behind it? H
1:21:36>> because a lot of them come with a
1:21:38disclaimer
1:21:40>> that it does not replace a you know a
1:21:43professional or a
1:21:44>> you know whatever pathologist evaluation
1:21:47or whatever disclaimer it might say but
1:21:50but my question is if they don't stand
1:21:52behind it then
1:21:56what are we doing like
1:21:59>> I would want
1:22:00>> for I would want somebody who said yes
1:22:02we stand behind it and if it Um if it
1:22:05gives an inaccurate diagnosis, we got
1:22:07you like
1:22:09>> right.
1:22:09>> Yeah. And and that's like I think my
1:22:12main because it's like we're in this
1:22:13like zone where it's like okay we can
1:22:16try it but if it fails and if I trusted
1:22:22a wrong answer they gave confidently or
1:22:24whatever. Maybe it gave a wrong answer
1:22:25confidently and I trusted it. Does that
1:22:28liability fall on the veterinarian? Who
1:22:30does it fall on? You know like we don't
1:22:32know.
1:22:33>> Interesting. Mhm.
1:22:34>> Yeah.
1:22:35>> Yeah. That's
1:22:36>> what's your feelings on like
1:22:39>> the ones that can read like basically
1:22:41take a whole bunch of pictures of the
1:22:43slide but then those aren't pictures are
1:22:45interpreted by a clinical pathologist.
1:22:48>> Those are great. Yeah. I mean that's
1:22:50that's great because um I mean digital
1:22:52there's a difference between digital
1:22:54cytology and AI.
1:22:56>> Um digital cytology so like some some
1:23:01products are a scanner.
1:23:03And um the scan of the slide can be sent
1:23:06to a pathologist like
1:23:09>> so you don't see if you have the
1:23:10physical slide or not being a path like
1:23:12you just
1:23:13>> there are some advantages to having the
1:23:15physical slide. So um as far as I know
1:23:18the current scanners that are available
1:23:21can scan up to um like a 40x objective.
1:23:25>> So you know like it's it's not oil it's
1:23:2940. And so I mean you can tell a lot of
1:23:31things at 40 and and for what I've seen
1:23:33of the images they're pretty good you
1:23:35know like but if you might be looking
1:23:38for like an organism or like bacteria or
1:23:42something like that that's that's kind
1:23:43of a limitation. I have not personally
1:23:46used the scanner myself so I don't know
1:23:51how difficult it is.
1:23:53>> Sure. So, you know, I I'm not sure on
1:23:56that, but I mean, you still have a human
1:23:59looking at it. So, and if they said, and
1:24:02there I believe there is the option like
1:24:04if they're like, "Hey, I'm concerned
1:24:05that there might be bacteria on here and
1:24:07I just can't see it." Like, you could
1:24:09still send the slide in to them in that
1:24:11case.
1:24:12>> Um, so, you know, there's options there.
1:24:14You're having a conversation with the
1:24:15pathologist. So, um, that's that's
1:24:18helpful. and the pathologist is probably
1:24:19going to tell you like I'm having
1:24:21trouble telling if I can see something
1:24:23because it's I I don't get that level of
1:24:25detail.
1:24:26>> Um
1:24:27>> but yeah, I think that's great. I mean
1:24:28that speeds up things. Um and you know
1:24:32for
1:24:34>> for the people especially if like
1:24:38>> you have to mail everything in or you
1:24:41kind of a rural practice, you know, it
1:24:43it probably would make things easier
1:24:45there. I I don't um I you know other
1:24:48than the the minor limitations that we
1:24:50talked about here, I I think that's
1:24:52fine.
1:24:53>> Yeah.
1:24:54>> Yeah. Well, um I have to ask before we
1:24:58because we've kept you a long time. I
1:24:59have to ask though before we go because
1:25:01Duckwall knows I'm a huge football fan
1:25:03and of course being an Auburn Tiger
1:25:05being SEC, you know, and you have on
1:25:07your resume that you referee college
1:25:09football games. What the Sam hell? How
1:25:12did you get into that? I got to know
1:25:13because I think that's so cool. And how
1:25:16did you end up loving football? Because
1:25:18I'm sure you probably do in order to
1:25:19referee it
1:25:21>> or appreciate it. So what's the scoop on
1:25:23that?
1:25:23>> I love football. Um so so I actually um
1:25:28so my dad went to Auburn and the funny
1:25:31part of this story is when I was in high
1:25:33school like they would go to all the
1:25:34home Auburn football games, my parents
1:25:36would.
1:25:36>> And um huge Auburn fans, my parents are.
1:25:40And so like they they went to all the
1:25:42football games and I as a high schooler
1:25:44could not care less. Like I was like
1:25:46>> don't care. Leave me be at home. Um and
1:25:49I went to a couple and I'm like it's
1:25:50hot, it's boring, like it's loud,
1:25:52whatever. And then I went to Auburn and
1:25:55was a student and got to sit in the
1:25:57student section
1:25:58>> and that changed everything. So, and it
1:26:01was also 2004,
1:26:04which anybody who's familiar with Auburn
1:26:06football history, that was an undefeated
1:26:08season
1:26:09>> with um you know, Cadillac Williams and
1:26:12some bigname players came out of that.
1:26:14So, it was an extremely exciting season.
1:26:17Um and so I was hooked at that point.
1:26:19Love football. Um and if with anything I
1:26:23do, I feel like it becomes an
1:26:25intellectual endeavor. So like like I
1:26:28love I you know I just enjoyed that
1:26:30first season. Then after that I'm like
1:26:31oh I want to learn more about the game.
1:26:33Let me learn about formations. Let me
1:26:35learn about like you know route patterns
1:26:39and like you know like and then I'm like
1:26:42oh football college football season's so
1:26:44short NFL goes even longer. let me start
1:26:47watching NFL games and learning their
1:26:49rules and like so anyway at some point I
1:26:52kind of like maximized what I could do
1:26:55from a spectator standpoint and I would
1:26:58be watching the games with friends and
1:27:00like I'd be like you know just watching
1:27:02a regular play on TV I'd be like oh
1:27:04that's a hold and they're like how'd you
1:27:07see that like I'm like I don't know like
1:27:10it just it's a hold it's right there and
1:27:11then you see the flag come out and then
1:27:13they're like holding on the offense and
1:27:15you're like Oh, okay. And so one of
1:27:18these days, one of my friends was like,
1:27:20"Maybe you should be an official." And I
1:27:22was like, "I don't even know if women
1:27:24can do that, but I will ask." [laughter]
1:27:26And so, um, so at that point, I uh
1:27:31there's the the way that you move up in
1:27:34football, like you you start at the
1:27:35youth level, you know, pretty much like
1:27:38anything. And so I was back at Auburn at
1:27:41that point and I just once again Google
1:27:44I was like how do I get into football
1:27:46officiating [laughter] and this is my
1:27:48location and I was like local high
1:27:50school football association um it's
1:27:52called the Big East football officiating
1:27:54association in Auburn and in kind of the
1:27:57surrounding areas. So I just emailed
1:27:59them and they were like yeah come on. I
1:28:02was like by the way I'm a woman I know
1:28:04that my name is kind of gender neutral
1:28:06so like [laughter] just letting you
1:28:08know. And they're like, "That's fine."
1:28:09And I'm like, "Okay, cool." And so I
1:28:11went to like my first association
1:28:13meeting and like literally had no idea
1:28:15what I was doing. And the first day we
1:28:17did like a rules test where like we had
1:28:19to like read a rules question and like
1:28:23give an answer. And like let me tell
1:28:26y'all, I thought I knew the rules.
1:28:29[laughter]
1:28:30You have no idea, like literally no idea
1:28:33what the football rules are until you
1:28:35have to become an official. Like you're
1:28:37like, "Oh yeah, 10 10 yards for a hold."
1:28:39You're like, "Yeah, from where?"
1:28:42>> Like, [laughter]
1:28:44>> "Yeah, it's like a previous spot spot of
1:28:46the foul." Like, you know, where is it?
1:28:48Where does it go? So anyway, yeah, I had
1:28:50no idea. And so I started officiating um
1:28:53youth that year. I didn't get in too
1:28:55many games. I was like a first year
1:28:56resident, so I did like one game on the
1:29:00clocks or something like that. They
1:29:01start you on the clocks and then you get
1:29:02on the field and then and then you move
1:29:04up from there. But um I went I started
1:29:08working it was like three years and then
1:29:10I I was on a varsity field that was
1:29:13pretty cool and I just was like okay how
1:29:16how does one go to college football like
1:29:20how would one do that and so there are
1:29:23these like essentially there are clinics
1:29:26or camps all around the country and so
1:29:30these clinics are meant to like you know
1:29:32you get um high level officials like NFL
1:29:36or um power four, power five, I don't
1:29:40know what we're at now. Um division one
1:29:42officials and they come to these camps
1:29:44and they give it's kind of like a CE
1:29:46conference, you know, like they give
1:29:49talks on like uh this is, you know, our
1:29:52past interference talk and you know,
1:29:54here's our holding talk and here's, you
1:29:56know, um God, there's so much that goes
1:29:59into it. What you want to do pregame,
1:30:00what you want to do, you know, pre- snap
1:30:02routine. That's like when what
1:30:04everything we do leading up to the play.
1:30:06Um so anyway, you go to these
1:30:10conferences and um and then there's also
1:30:14a lot of them have time to be on the
1:30:16field. So they get like a local high
1:30:18school or something and you go out there
1:30:20and you have people clinicians standing
1:30:22behind you watching you work and and
1:30:24telling you how to improve. And so I
1:30:27went to my first one in Arkansas. Um,
1:30:31and at the time I just thought I was
1:30:35going to like learn more about football
1:30:39or whatever. Like I thought that's what
1:30:40they were. I thought they were CE
1:30:41conferences, you know. Um, what I didn't
1:30:44know was that there's scouts from like
1:30:47all divisions there watching you and
1:30:50like Anyway, so so I get there and the
1:30:54the guy in charge of it is standing up
1:30:55at the front and it's like the the
1:30:57opening um the opening meeting and he
1:31:01was like he was like, "All right,
1:31:03everybody, I just the main goal of us
1:31:06being here is to get better." Right?
1:31:09Right. Get better.
1:31:12not to get hired, it's to get better. If
1:31:15you get hired, that's great, but it's to
1:31:16get better. And he just kept saying that
1:31:18and I was like, I thought that's why we
1:31:19were here. Like, what are we talking
1:31:21about? [laughter]
1:31:23Like, like I somebody filled me in. They
1:31:26were like, oh yeah, that's a scout.
1:31:27That's a scout. That's a scout. I was
1:31:29like, oh my god. I'm like, I don't know
1:31:31what I'm doing. So, um, and I was also
1:31:34at a position I had never worked before
1:31:37because that was the only one left when
1:31:38I signed up. So, I'm like literally like
1:31:41flying blind. Um, but I was there and I
1:31:44did it and um I did end up getting
1:31:47picked up by a division 2 conference.
1:31:50Um, and I got to work like clocks that
1:31:52first year and then eventually worked my
1:31:54way onto the field. Um, but yeah, so um
1:31:59so it's it's a ton of fun. Uh it is also
1:32:03like of the things I've done, it has
1:32:07been the hardest thing for me to learn,
1:32:11for me to become proficient at. Uh the
1:32:14steepest learning curve, like it's so
1:32:19hard. Um because everything happens so
1:32:23fast
1:32:25>> like when you're standing on the field
1:32:27>> and you're at like you are literally eye
1:32:30level with the player. I mean like
1:32:31they're like up here but you know like
1:32:33you're eye level with everything and
1:32:35there's not many people know this but on
1:32:37a college field there's eight of us on
1:32:39the field like you know people they only
1:32:42see the the one guy that makes all the
1:32:43announcements the head referee but
1:32:45there's there's seven others out there.
1:32:48Um, and it's like the crew communication
1:32:52like that is your family when you're on
1:32:54that field and like you know most of us
1:32:57it's like you're assigned a crew at the
1:32:58beginning of the year and that's who
1:32:59you're with throughout the entire year.
1:33:01And I have had some fantastic crews and
1:33:04it's like that that's your family for
1:33:06the fall
1:33:07>> and just you know it's
1:33:11>> it's amazing. I love it so much. Um it
1:33:15has taught me so much has developed me
1:33:17as a person.
1:33:19>> Um I you know like you you can't be
1:33:27you you have to be confident.
1:33:30>> You have to be confident. You cannot
1:33:33look unconfident.
1:33:36Like it's, you know, like I mean imagine
1:33:39what what it would look like if you
1:33:40looked at an official and they looked
1:33:42like they didn't know what they had, you
1:33:43know, like you were like, "Oh,
1:33:45>> yeah." Like you have to be confident and
1:33:47that means you have to prepare. You have
1:33:50to know the rules. You have to have the
1:33:52experience. You have, you know, like you
1:33:54have to be confident to walk on that
1:33:55field. And the other thing is you like
1:33:59coaches will ask you questions sometimes
1:34:02very forcefully but they will ask you
1:34:04questions all the time and like you just
1:34:07you
1:34:08>> you cannot be defensive like
1:34:12>> you know you you have to learn because
1:34:14they might yell things at you that are
1:34:16like
1:34:17>> you know not very nice
1:34:19>> and but like you have to be the
1:34:22professional you have to be not
1:34:24defensive. So, like, and that was an
1:34:26art. That was the hardest part for me
1:34:28when I first got started was having a
1:34:31coach yelling at me, you know, like
1:34:34they're they're just like,
1:34:36>> "You see that?" You know, like, and
1:34:40you're like,
1:34:41>> you you just, you know, it's it's an art
1:34:43learning how to handle that. And now
1:34:46like the the beauty of it is once I once
1:34:49I got enough advice. I have a lot of I
1:34:52have one very good mentor but I have a
1:34:54lot of others that I get advice from and
1:34:57all of them have told me like most of
1:35:00the time all they want is information.
1:35:03>> So if you just give them information
1:35:05like they it usually it's so helpful.
1:35:09Like I mean they just like I remember
1:35:12one time um you know I'm I'm deep. I'm
1:35:16way back behind the defense like looking
1:35:18for all the pass interference calls and
1:35:20the catch no catch and sideline stuff
1:35:22you know like um that's where I am. So
1:35:25I'm essentially like 35 40 yards away
1:35:29from the quarterback, the line of
1:35:30scrimmage, all that. And I remember one
1:35:32time um a coach was standing next to me.
1:35:36um they were kind of backed up in their
1:35:38own uh in their close to their own uh
1:35:41goal line and so the coach is standing
1:35:44next to me and it's like a long pass but
1:35:47it's incomplete and then like I have to
1:35:49follow that play with my receiver and
1:35:51then I come back and the coach is like
1:35:54there's a face mask there's a face mask
1:35:56on the quarterback and I'm like okay
1:35:59like I'm 48 you know I'm like okay and
1:36:02then I'm like he's like I can't believe
1:36:04you haven't seen that you like you're
1:36:06like going crazy. And I was like,
1:36:07"Coach, coach, coach, hang on, hang on.
1:36:10How about this?" I'm like, "We don't
1:36:12have a flag on the ground. How about
1:36:13this? At the next dead ball period, I
1:36:16will go ask them what they saw and I
1:36:18will make sure that they they saw it. If
1:36:21they didn't see it, I'll tell them what
1:36:23you said. If they did see it and they
1:36:24have information, I'll bring it back to
1:36:25you." And he said, "Okay, that sounds
1:36:27great." [laughter] It was like, "Okay."
1:36:29And then it was like right after that
1:36:30there was a timeout called and I ran up
1:36:32there and I was like, "Hey guys, coach
1:36:35is yelling about a face pad." And they
1:36:36were like, "No, no, no." They he got it
1:36:38was like a one of these where they got
1:36:39the shoulder pad and it looked like a
1:36:41face mask from, you know, 40 yards away.
1:36:44And I remember I uh I was they're
1:36:48they're doing this, you know. No, he got
1:36:50his shoulder pad. And I hear the coach
1:36:53all the way on the sideline that was no
1:36:55shoulder pad. There was no I was like
1:36:58that's so funny. And then I I ran back
1:37:00to I was like look coach that's what
1:37:02they say. He's like all right you know
1:37:04that's fine but he's like I gota I gotta
1:37:06fight for my team. I'm like I hear you
1:37:08coach. Like that's great you know like
1:37:10I'm I'm with you. and he's and we were
1:37:12good after that, you know, but it's just
1:37:15so funny because like most of the time
1:37:16they just want information and as long
1:37:19as you can get information to them, um,
1:37:22you know, it clears up a lot of
1:37:23situations and you just you just have to
1:37:25have thick skin, you know,
1:37:27>> they're yelling at the stripe shirt,
1:37:28they're not yelling at you personally.
1:37:30So,
1:37:31>> and that that's how veterary medicine
1:37:33is. So,
1:37:33>> I was going to say there's so many
1:37:35parallels
1:37:37to the owner and their patience.
1:37:39>> Yes, it is. There's so many parallels.
1:37:42Like people just want information and
1:37:45>> Yeah. And it's like you can't you can't
1:37:47take it personally and sometimes really
1:37:49hard not to take it personally, you
1:37:51know, like but
1:37:52>> we've all been there. We've all been
1:37:53there.
1:37:54>> We've all been there.
1:37:55>> Yeah, it is hard. I mean, Duckwell and I
1:37:57have had conversations about that. So,
1:37:59you know, where it's like, you know,
1:38:01it's hard sometimes not to get your
1:38:03feelings hurt, you know, in those
1:38:04situations. So, it's
1:38:07>> Yeah. hard lessons to learn sometimes.
1:38:10Well, listen, we want to
1:38:12>> Yeah. Thank you so much for taking the
1:38:14time to like let us learn about you and
1:38:16your life, but also pick your brain
1:38:18about clinical pathology and AI and
1:38:21wherever AI is going. I guess we're all
1:38:24along for the ride and we'll find out.
1:38:26But
1:38:28>> yeah, we can't stop it. So,
1:38:30>> that's true.
1:38:31>> I don't know. I don't know if you've
1:38:32looked, but we try to break the wheel on
1:38:34this podcast, so you never know.
1:38:37[laughter]
1:38:38>> Yeah. Yeah. It's like we can't stop it,
1:38:40but we we can know what information we
1:38:43need and what questions we need to ask.
1:38:45So,
1:38:45>> Exactly. Exactly.
1:38:46>> And and ultimately, I think it comes
1:38:49down to we need to decide
1:38:51what do we want out of it,
1:38:53>> you know? What where can it be useful
1:38:56and like time will tell, right? Because
1:38:58like we don't know how far it's going to
1:39:00go, but what what do like as a society,
1:39:04I feel like we need to ask what do we
1:39:07want? Where can this be useful? I I
1:39:11remember I saw something that was like
1:39:15>> what if we used it to make ours or our
1:39:19employees lives easier rather than
1:39:21trying to replace people.
1:39:23>> Yes.
1:39:23>> You know, like I think that's a question
1:39:25we really need to ask because
1:39:28>> it feels like there's
1:39:30>> there's you know, depending on where
1:39:32you're sitting at the table, there's a
1:39:34pressure to Yes, I want to use it. I
1:39:36want, you know, my my life to be um more
1:39:39efficient, less mundane, busy work,
1:39:41notes, things like that. Like obviously
1:39:43it it could very easily help there
1:39:46versus like on another side of the table
1:39:48might be thinking like how do we replace
1:39:51people with this? And I I
1:39:53>> I really think we need to stop and ask
1:39:56the critical questions like is that
1:39:58really what we want? And if we do, what
1:40:02are the unintended consequences of that
1:40:06decision?
1:40:07>> Yeah.
1:40:07>> So anyway, those are my thoughts. My
1:40:10bills.
1:40:11>> Great point. Great point.
1:40:12>> Absolutely.
1:40:13>> We thank you. We appreciate you for your
1:40:14thoughts. And
1:40:15>> thank you for having me on. This was
1:40:17super fun.
1:40:18>> Well, you're always welcome back. So,
1:40:20we'll just look ahead here.
1:40:22>> And y'all can come to Auburn anytime.
1:40:24>> I love Auburn. I've been there a couple
1:40:26times. I love it.
1:40:27>> Go to a football game.
1:40:29Yeah, I will unfortunately never be
1:40:31allowed to work an Auburn football game
1:40:32because I have obviously way too many
1:40:34ties with it. But but I can get you into
1:40:36a football game and you know on a free
1:40:38weekend I could definitely go with you
1:40:40too. [laughter]
1:40:41>> That sounds great.
1:40:43>> I can tell you all the intricacies about
1:40:45what's going on that you have you know
1:40:46like behind the scenes.
1:40:48>> I love it. I love it. [laughter]
1:40:50>> All right. Thank you. Thank you so much
1:40:52for tuning in to another episode of the
1:40:53veterary round table. Remember send in
1:40:55those questions and leave us a review.
1:40:57If you enjoyed this episode or previous
1:40:58one, we'll see you next week for another
1:41:00episode of the veterary round

About this transcript

This page contains the full transcript of Can We Trust AI in Veterinary Diagnostics? w/ Dr. Jordan Towns | The Veterinary Roundtable - EP189 by The Veterinary Roundtable, generated from the public captions YouTube serves with the video. The transcript has 17,918 words across 2,592 segments, with the original timestamps preserved so you can click any line to jump to that moment in the embedded player.

What you can do with it

Use the transcript to take notes, quote the speaker, build a study guide, generate a summary with ChatGPT or Claude via the YouTube Summary tool, or export it as a timed subtitle file with YouTube to SRT. You can also re-open it in the transcriber to translate the transcript into 100+ languages.

Free YouTube transcript tool

YouTube2Text is a free YouTube transcript generator — no signup, no daily limit. Paste any YouTube link and get the full transcript instantly, with timestamps, click-to-jump, translation to 100+ languages, AI prompts for ChatGPT, Claude, and Gemini, and exports to TXT, SRT, VTT, or Markdown.