📷 Mariam Said
Journalist: Amy Stewart
Amy: Hello and welcome back to SciSection! I'm your journalist Amy Stewart, for the SciSection radio show broadcasted on CFMU 93.3 FM radio station. We are here today with Dr. Mark Walker, a high-risk obstetrician, a clinical epidemiologist, and the Vice-Dean of Internationalization and Global Health at the University of Ottawa. Thank you so much for joining us today, Dr. Walker.
Dr. Walker: Thanks very much, Amy. It's my pleasure to be here.
Amy: Alright, to get us started, why don't you give us a briefing of your educational and career background and how you got started in your field?
Dr. Walker: Great. So, I started off my career at Queen’s University, I did two years of undergrad and then went straight into medical school. Prior to that I worked part time as a computer programmer, I think that’s relevant to this conversation. But I did a lot of coding. I finished med school in ‘93, came and did my residency in obstetrics and gynecology here at the University of Ottawa. Then I did a fellowship in internal fetal medicine or high risk obstetrics in Toronto for two years and also completed a masters in epidemiology at that time. Started in the faculty here in 2000 and as a clinician scientist. Went back to school in 2010 and did a MHA at Harvard to help me with my leadership skills and managing lots of projects. Since being here I’ve established two really large research groups, the OMNI research group as well as, it's not a research group but it's a clinical registry, BORN Ontario. And was chairman of obstetrics and gynecology for seven and half years before becoming Vice Dean in Internationalization and Global Health.
Amy: That's a pretty awesome career so far! I love how you've taken on such an important leadership role in the healthcare community especially at uOttawa but I’m sure it has outreaching effects all over Canada.
Dr. Walker: Yeah, I've been very lucky, I’ve had a career where my work’s my hobby and some days I could say I’ve never worked a day in my life.
Amy: So you alluded to having history working in coding and with computers, and you also recently led and published a study that uses deep learning A.I in fetal ultrasounds to diagnose birth defects, can you tell us how you got on that project and kind of explain the methodology behind it?
Dr. Walker: Sure. So I've always had an interest in computers and I think my success in research is actually data management, so using large data sets and developing databases, and I’m quite proud of all the work actually but particularly BORN Ontario, which is arguably the best prenatal registry in the world. We do fantastic work and have high impact publications. I started to understand more and more about A.I as it relates in the business world and then I decided to learn more so I read some books, very high level on it. Then I took a course at MIT remotely, just to understand the language more than anything else. And then, I saw its power and I think the two ways that it's been interdigitated in health research and medicine: one is in imaging. It's extremely good at image recognition and arguably equal to humans, you know physicians and if not better, because of less bias. And the other area is predictive modeling algorithms, around disease prediction, response, and treatment.
Amy: That is very cool. Can you tell us about some of the findings of that recent study you did and how does the A.I exactly diagnose those birth defects? Is it just by looking at images? And how does it compare to a human’s eye or a physician diagnosing, themselves?
Dr. Walker: So artificial intelligence and deep neural networks have very much been developed to work similar to neuron function. And so a traditional computer programming it’s all logic statements: if this happens and you do that, or not statements. It follows basically a recipe if you will of how they execute the programs.
Where deep learning and neural networks differ is: at any one signal node like the neuron, it can have multiple synapses coming in and in this case it’s multiple imputations into silicon. And once it reaches that critical mass and it fires. So it was developed loosely in the architecture of the occipital lobe of the brain and that explains why it's so good at image recognition.
One of the challenges in my field is picking up fetal anomalies and when they’re picked up early, the more choice we can give parents as well as the possibility for surgical intervention. And often they come to us late. And then there's the other issues, you know, a large portion of the world doesn’t have access to any imaging. So with our research group is “hey why don’t we start to catalogue and show the efficacy of artificial intelligence in interpreting fetal ultrasounds so that one we can prove the quality in our own country but also create a system that can be used in lower income countries.” Loosely given it the name OBUS cloud - so OB ultrasound cloud. And the first part of the study we did was sort of proof of principle, just working at the work flows, like do you get an image off of the PAC system to where do you store it, which deep learning algorithm do you use? So it was really a proof of concept paper and it was looking at a specific anomaly called cystic hygroma which you see early in pregnancy. And the limited data set that we had like 300 images, the computer, it’s deep learning was incredible. There area under the curve of almost perfect, there was only a couple mistakes. And the other beauty of it is you don’t code anything in, you just show it an image, oh it’s a cat, show it a dog, other image is a tree and you feed it more and more images and then you give it an image it hasn't seen before and it says cat, dog, or tree and that's the accuracy. And so it worked very very nicely. And then we’d like to get to is in low fidelity, you have an iphone and you have a probe that bluetooths to it, you scan the fetus, it goes up to the cloud and then it says normal or abnormal, gestational age, placental location. And it would be a huge breakthrough here but also in low income countries,
Amy: That is fascinating technology, I love how you mentioned how it can be used in these low income countries because like you say they may not be able to staff a whole research team or someone to diagnose problems like that, so having access to tools where they can use data from other scientists is fascinating and I feel like that works perfectly with your role as the Vice Dean in Internationalization and Global Health Care. That's great to see you connecting the two like that.
Dr. Walker: Yeah thank you, it is exciting and it's a lot of hard work but it's the passion and the ability to impact large populations, particularly those that aren’t as privileged as ours.
Amy: That is amazing. So I also wanted to talk a little bit more about your role as the Vice Dean in Internationalization and Global Health Care and what that means and what kind of projects you guys are working on at uOttawa.
Dr. Walker: So I'll separate it into both, there's an internationalization stream which is mostly University to University, kind of north-north collaborations. So for example, we have collaborations, there are news ones they just got back from Cambridge, one in neuroscience, it is very strong and both institutions. Cancer with the top cancer institute in France, at University de Paris-Saclay. That’s the internationalization piece and we're really looking for student exchange, student mobility to increase research capacity. The global health I’m quite proud of the work that we’ve done there. We have an amazing assistant dean, Dr. Manisha Kulkarni, and we have projects. We've decided to conform to best practices in global health and try to go deep in a few places as opposed to being spread out and in alot. And train physicians, train the trainer and really meet the needs where the country is at and what they need. So our team is actually going to Tanzania. It's the Christian medical center, it's a university and a hospital. The big area of need, I mean it's everything but it;s surgery, obstetrics, anesthesia, and paramedical. And so we have two obstetricians going, we have two general surgeons, we have an anesthetist, public health people as well. And the idea is to, especially now virtual training, is to help them develop capacity in training for their own physicians both at the undergraduate level but particularly at the post-graduate level. And right now in Tanzania, it's just one example. The number set up by the WHO for surgeons, obstetricians, and anesthetists per 100 000 is 20, and they have less than 1. So the ability to have a large impact and a lot of exchange of students on both sides so bilateral mobility, bilateral mobility of professors and then ideally develop the most sustainable curriculum that we can share and that different people take different parts of it. And then we also have in Franco-Africa we have a project that Dr. Marie-Hélène Chomienne has led for several years in Benin and actually we were on the phone with them this morning, were doing research together building capacity. And right now there's a very strong family medicine program, one of our pediatricians and neonatologist, Nicole Rouvinez Bouali, set up a phenomenal program on neonatal resuscitation and it’s been adopted through the whole country. And we are going to start to have obstetricians as well, and be involved in training their physicians.
Amy: That is some really amazing work, I had no idea all of that was going on just from uOttawa. I really like what you said about how the projects you guys work are more about focusing on what each country or region really needs as opposed to trying to cover everything very thinly spread and this way focusing your resources more specifically, I think that’s a very good tactic. And I love to hear how uOttawa’s been helping out the international community. For my last question, I would like to ask what kind of advice you'd like to give to undergraduate students who want to pursue a career in medicine and research?
Dr. Walker: So for both of them, I have to say from my career path I never thought I would end up at obstetrics, I didn't expect to be an academic. So sometimes you follow the path through the forest and it takes you places you haven't known or expected. But
I think the most important thing is choose something that you love, like I said for myself, I’ve really not had to work very much in my life because everything I've done has been my passion.
I’m a bit of a pied piper when it comes to obstetrics just because for me, when I saw my first baby born it was truly a miracle, nothing inside could explain it. It still has that magic after 20+ years of practice. And then for the research I think, as a physician or surgeon you can help one patient at a time which is amazing, as a researcher you can help thousands and possibly hundreds of thousands. And there is still so much we don’t know, so much we can diagnose and we can’t treat.
And to be honest there’s never been a time that's more exciting in science and discovery than in the last five years. As a physician scientist you have the gift and the privilege of being able to ask and answer those questions that you can’t solve in daily practice and be part of that process.
And so it's a wonderful career path and you never stop learning and it's always a new challenge. So I don't want to ever say not working because I love it so much and just like artificial intelligence we talked in the beginning, that's gonna change the way that we practice and there will be fellowships in informatics and A.I and a lot of care is going to be driven by algorithms and it’s not going to replace us but it is going to make our life much easier.
Amy: Thank you for the wise words. I think you make a great role model, you're taking on so many different aspects of healthcare from the leadership roles to the actual working with patients and research and all that and to see how passionate you and it doesn’t even feel like work, I think is very inspiring who feels like thier working their but off to get themselves to medical school or into their PhD program or whatever.
Dr. Walker: And in fact I didn't enjoy medical school much because it was hard work. But once you find what you love, then you give it a hunger that's never satisfied, which is good, they call it the fire of the belly.
Amy: Well thank you so much for joining us today Dr. Walker. I had a great time interviewing you, and I loved hearing about all of the projects you've been working on and everything that's been going on at uOttawa that I feel like a lot of the student body doesn't know about, I mean the outreach you guys do is amazing and I'm excited to see where you take A.I in your career and all of the applications it can have beyond that.
Dr. Walker: Yeah, it is going to be exciting and exciting for a lot of us, we’re a big team and a lot of people are interested like fellows and my colleagues and it’s ince to be early into the field because there's not a lot of people there yet. Somebody once told me “you don't have to be the best, just the first.” Thank you very much.
Amy: Yeah thank you, for joining us. That's it for this week of SciSeciton! Make sure to check out our podcasts available on global platforms for our latest interviews.