Today’s Advertiser helped make the production and delivery of this episode possible. But the themes expressed by the guests do not necessarily reflect the views of Inside Reproductive Health, nor of the Advertiser. The Advertiser does not have editorial control over the content of this episode, nor does the Advertiser's sponsorship constitute an endorsement of the guest or their organization. The guest's appearance is not an endorsement of the Advertiser.
Lab automation in IVF is no longer theoretical, it’s been proven.
Proof of concept doesn’t mean it’s ready to replace embryologists…
But it does mean this works.
Chief Scientific Officers Jason Barritt of Kindbody and Jacques Cohen of Conceivable Life Sciences join the episode to discuss a recent study published in Human Reproduction examining AURA, the robotic lab system developed by Conceivable Life Sciences.
We dive into:
What “proof of concept” actually means in IVF lab automation
Why this study matters (And where it falls short of current standards)
The role of automation as a testing ground for new lab technologies
What a fully automated IVF lab could unlock
Whether “hub and spoke” models in fertility have been misunderstood (and what they could actually become)
If automation continues to progress, the scale of what’s possible in fertility care may look very different than it does today.
A Historic First in IVF: Can Day 0 Be Fully Automated?
For years, IVF automation has focused on single-point solutions. One step. One tool. One task at a time. Human Reproduction recently published Conceivable Life Sciences’ Day 0 research to answer a much bigger question:
Can multiple automated systems sequentially perform Day 0 IVF procedures?
This is the first published data exploring whether integrated automation can execute the earliest phases of IVF, from retrieval forward, as a coordinated system.
Inside the Day 0 paper, you’ll discover:
Why Day 0 has remained one of the least standardized stages in IVF
How sequential automated systems were engineered to work together—not in isolation
What technical validation data reveal about system performance
How human oversight is integrated at every stage
Why this marks a shift from single-point tools to workflow-level automation
This isn’t about replacing embryologists. It’s about proving whether complex IVF procedures can be supported by coordinated systems designed to deliver consistent, expert-level performance.
Before this paper, there were proof points. Now there is system-level evidence.
If you care about the future scalability of IVF, this is required reading.
👉 Conceivable Is The Ultimate Family Business, follow on LinkedIn.
-
Dr. Jason Barritt (00:00)
It is amazing what automation and AI has been able to somewhat replicate us humans. Because the truth is when you sit down at one of these apparatus and you start working at it, your brain is functioning on so many levels, with so many things and so many axes and so many focal planes that now imagine you're trying to let a computer learn how to do that and then manipulate it with electronics in a timely fashion that does not harm humans.
Griffin Jones (00:38)
Proven. Lab automation as a concept is proven. That's according to the findings of a paper published last year in Human Reproduction and the conclusions of my guests, doctors Jacques Cohen and Jason Barritt Now, proof of concept does not mean non-inferiority. No one's saying that. It just means that this thing can work. Inside Reproductive Health is not a medical journal. We don't do peer reviews here. And Cohen and Barritt try to very clearly separate the findings of the study
from future speculation. And I take us back and forth between the two because it's my show. Conceivable is not responsible for that. Neither are Cohen and Barritt. I like to shoot the breeze. If you want a medical journal, read Human Reproduction. And you should, by the way, because we'll link to that study.
in the page where you find this podcast episode and the email that it goes out in.
And you can easily find it on Conceivable Life Sciences website.
So find and read this paper in the appropriate place. Here's what stood out to me. In the study that analyzed Conceivable Life Sciences robotic lab automation system, AURA five healthy babies were born, 64.3% of the eggs fertilized
And out of hundreds of eggs, none were damaged by the robot.
Can your embryologist say the same?
Dr. Jason Barritt is, of course, the chief scientific officer of KindBody, and Dr. Jacques Cohen is the CSO of Conceivable. While they're both very tempered about their excitement of the study, still not meeting or exceeding current standards, it proves the concept of lab automation as a viable possibility.
And as Dr. Cohen observes, a fully automated IVF lab would be the supreme testing ground for new solutions in the IVF lab. These observations allow us to think about the future. If the words hub and spoke excite you in the fertility space, the second half of this episode is for you. We don't know what hub and spoke means right now as we misuse those words all the time in this field.
Now, when you listen to this, you will see what real hub and spoke looks like, what IVF will look like in some years time at a magnitude that so far people have only dreamed.
I hope you enjoy this conversation as much as I did.
Griffin Jones (03:47)
Drs. Cohen and Barritt, Jacques and Jason, welcome back to both of you to the Inside Reproductive Health Podcast.
Dr. Jacques Cohen (03:55)
Yeah, good to see you, Griffin.
Dr. Jason Barritt (03:56)
Thank you.
Griffin Jones (03:57)
Jacques is the concept of IVF lab automation proven.
Dr. Jacques Cohen (04:02)
I call that an A question. I prefer B questions. That's a really, really good question. think Jason will be probably in a better place and more objective to answer that. think we're going in that direction. So proven, we'll need a lot more time for that, but we're going definitely in that direction.
Griffin Jones (04:21)
What's your answer, Jason?
Dr. Jason Barritt (04:21)
I think I'm willing to go a little farther than Jacques did, to be honest. I think one of the main items that we're going to discuss today is actually proven that we have reached it. I am not saying, and I want to be very careful.
I am not saying it's fully ready for full prime time. It's not better than what we can offer in some ways. But that doesn't mean it's not proven it can work. That, I think we've reached.
Griffin Jones (04:50)
So before I go into why you feel that way, Jason Jacques, what causes you to maybe stop a little bit short of that?
Dr. Jacques Cohen (04:58)
Well, I look at proof in science and medicine in particular, we look at proof is a very loaded word, very loaded terminology. And so I look at it from an evidence-based medicine point of view, which is usually a process that takes many years, And so I was answering from that perspective. So are we in the right way? there's no doubt about it. There's no doubt about it. We're moving right ahead. ⁓
Griffin Jones (05:05)
Mm-hmm.
Dr. Jacques Cohen (05:23)
A lot of good papers have come out in the last year or two. The paper I think you're interested in was published at the end of December by our group, Conceivable Life Sciences. And that work I think really is very interesting. We were doing a proof of concept study, in actual fact started in 2023 and finished in 2024.
on replacing single vitrified blastocysyts which were obtained after going through a series of different automation steps, not just one, but several. And I think what Jason is alluding to probably is the fact that it's not just, we not singled out a single thing, we did several. And this combined two or three or four automation steps that we can separate.
when you handle eggs and finding the eggs automatically and then processing the eggs so that they're ready for ICSI then that those are really, I think we would agree there are two steps. Some would consider that one thing. I think it's like really two distinct steps with greatly different outcomes per embryologist. They're very sensitive steps. The other one was sperm preparation. We don't talk much about sperm preparation. All technology that's been
come a little better over the years. But we have automated that at least the first attempts of it. And then there is ICSI, which we have automated and have gone into autonomy of the different steps that involve the ICSI process, of which we think there are 15 to 17 steps. And so that whole thing, we have tried to combine in a very small set of patients, because you have to be very careful when you do this.
for the first time. So you can't go kong ho. It's just a randomized clinical trial, obviously IRB reviewed in Mexico, in Guadalajara, with the HOPE IVF clinic. And that was published in December. Five babies born out of nine, no, 11 transfers. Oh my God, Jason, help me out. 11 patients, 12 transfers or 12 attempts here.
Dr. Jason Barritt (07:37)
Yeah, nine
pregnancies and five healthy births.
Dr. Jacques Cohen (07:41)
Yeah, five births, five babies born.
Griffin Jones (07:44)
So you do make a good point, Jacques, that I should define terms and by, if our measure is automated lab technology producing the same or better outcomes as today, that would be beyond proof of concept in my view. By proof of concept, I simply mean this can work, that it proves that it can work. So you would agree that the concept has been proven in that sense.
Dr. Jacques Cohen (08:01)
Yeah. yeah, yeah.
absolutely. absolutely.
Yeah, this can work. For instance, fertilization rate. Well, would we have liked to see it higher? Yes. But it's on the way. It's on the way. I think an important thing people need to realize, yeah, it's a concept study, but we haven't excluded anybody. We haven't done 50 patients or 500 embryos before this, before we went into this study.
This includes all the patients. This is the learning curve that we have published in the form of a trial that makes sure that the patients have a good chance by taking about half the eggs and handling those through the manual laboratory, the regular embryology laboratory, because we were doing this in an embryology laboratory. This was not done in a separate unit. And so half the eggs were treated conventionally, regular IVF, ICSI and...
and the other half were handled by us. So that's how this was done and it includes all the data.
Dr. Jason Barritt (09:08)
So that's what I want to jump off here, Griffin. So ⁓ the amazing thing is I was trying to do the calculation last week. I was a brand new PhD, just finishing up presenting my work when I was thankfully plucked out of the anonymity of doing my stuff on the side. And I was given the opportunity to go work with Jacques Cohen about 27 years ago. And I came in and did
a three-year fellowship there. And yes, under his leadership and the things that were done there, it is important to get to the science of the proof. It is not just about what you might be able to do one time and things like that. You gotta prove it. And you gotta actually have somebody else prove that you can do it too in order to actually be able to put this out and to put it into a paper. So I wanna fully
my director here, one of my absolutely most important things in making sure that I had a great career and I started off wonderfully. I definitely appreciate that. But what it is is the idea of what we really need to discuss is the paper is it is a proof of concept and it's the learning curve as Jacques said. The thing is if you're willing to talk about your learning curve and put it into a paper and get it published, even when it's not perfection, that demonstrates how
Dr. Jacques Cohen (09:55)
Okay. Okay.
Dr. Jason Barritt (10:24)
really important this step is because no one just immediately jumps to a solution and proof. This is many, many stages and many steps. I'll say, technically this work is started and or not even mostly completed nearly three years ago in its preparation. We are three years past this already, but this was the first time they're putting it out.
Dr. Jacques Cohen (10:38)
Okay. Okay.
Dr. Jason Barritt (10:51)
in an organized way so that everybody else can look at it and analyze it and that is the key thing here. So he was cautious about calling it proof, but I am saying they've reached that point at this point with all respect to my mentor. I think they have reached that with this paper and that's why it's so important. As he said though, is it perfection? It is not. It's not bad at all and that's what really needs to come out of this.
Dr. Jacques Cohen (11:15)
Yeah.
Dr. Jason Barritt (11:18)
It is amazing what automation and AI has been able to somewhat replicate us humans. And that is the key factor of what is the outcome of this paper. And as Jacques said, automating ICSI is not 50 steps, there's 500 steps. Because the truth is when you sit down at one of these apparatus and you start working at it, your brain is functioning on so many levels.
with so many things and so many axes and so many focal planes that now imagine you're trying to let a computer learn how to do that and then manipulate it with electronics in a timely fashion that does not harm humans.
Wow. That is an amazing thing. And it is not easy to do. So this is a heck of a proof of concept and beautiful learning curve paper.
Dr. Jacques Cohen (12:01)
Yeah. Yeah.
Griffin Jones (12:08)
Did you, as someone that doesn't know how to read scientific literature, Jason, did you find the paper lacking for anything that you would have wanted to see validated?
Dr. Jason Barritt (12:18)
So the answer straight up is no. would I love them have potentially jumped over all 10 hurdles and run the entire Olympic race and won it in gold medal? Yeah, would've loved it. That's not how you learn to race. Right now, we're learning to race. This was the demonstration of that. So I don't think it's lacking anything or
Dr. Jacques Cohen (12:34)
Yes.
Dr. Jason Barritt (12:38)
not showing us anything that I would expect it to have. Would I always want it to go farther? Yes. But guess what? They've been working on things for three more years since this. Guess where they might be now with this. This was the true proof all the way through birth and healthy children. That is a giant step and a lot of time to prove that automation can work in the field. So I'm not saying there's anything lacking. It's just like you always, you always want to
Dr. Jacques Cohen (12:58)
Okay.
Dr. Jason Barritt (13:06)
win everything right away, don't you?
Griffin Jones (13:09)
So Jacque said the paper was published in December. What were the dates of
the RCT?
Dr. Jacques Cohen (13:15)
The RCT started in late 2023, December 2023, was finished around April 2024. But then you need to complete it because all the blasts were fictified. That's the concept. We go with what the host clinic normally does. And that's what they do. They fictify everything. And then the patient comes back two, three, four.
cycles later. In our case, because they had separate consented, they were asked to come back and not wait years. So they and they agreed with that process. They came back as soon as they were ready and the clinic was ready. And that means that the premises then come in, in the course of the next months. So you're always a little behind. So that's why there's such a long period and gap. Yes, of course. And then of course, nowadays,
Very difficult enough for people to publish something without having live births. So you have to wait even longer than you normally would want. So yeah, a lot of patience was needed to reach that point. That's why it only saw the light at of the last week of December. 2025.
Dr. Jason Barritt (14:23)
So additionally, remember, they
had to develop all this technology before they even jumped off to do the first patient. So that's why this is a multi, multi-year process here to even get to actually applying it clinical. This is not theoretical. They did it on humans, made babies. They're now here. That is a huge giant leap. And that's what this talk's about.
Dr. Jacques Cohen (14:44)
Yeah.
So we invaded somebody's IVF lab and took all their equipment and made them miserable for about half a year or longer. I'm very grateful to the group at Guadalajara at the time. And nowadays we do this, the next step which you're asking about in Mexico City because
Griffin Jones (14:48)
What comes next,
Dr. Jacques Cohen (15:10)
It is something to evade your lab. know, Jason knows this very well. We're very wary about visitors. And now they had visitors who were invasive visitors, people who were coming in with computers, know, ⁓ took over entire stations and tried to automate those stations and cameras and microphones. They brought in their mobiles. They were doing things from their mobiles.
Griffin Jones (15:28)
and cameras and microphones.
Dr. Jacques Cohen (15:39)
⁓ the thing was completely digitally controlled. So you didn't really have to be there, but in order to look at a system that was makeshift, the systems engineers had to be there. And ⁓ software engineers at least wanted to be close by. So it was very invasive. So that's already very remarkable. What's also remarkable, we didn't realize that in the beginning so much that the whole thing is actually not that we're sitting at the microscope and then moving things.
directly from each movable device, from each smart device that you can control. No, the whole thing went just on a computer, from a computer, not so much from a phone, that would be very risky, I think, but from a computer using your keypad, using your mouse, and then you could direct all the steps that were taking place. Now, we published a paper earlier last year where we...
where we showed that you could take that over thousands of miles. So the desk where I'm standing right now was also used, I used that for a few acts to do the ICSI process digitally, giving it commands to do some autonomous steps, but that was so early in the process that a lot of times those then fail, but that still means you have to digitally control things. You know, when we do micro-application and ICSI, we're sitting there with these joysticks, right? You must have seen this in pictures and in action.
It is like driving your car without a steering wheel, but you have two joysticks, a bit like flying in some cases. Okay, so you have all these other controls for suction control, aspiration, taking a sperm in a needle, releasing it, holding the egg on another device. So you have all these little controls. Yeah, there you go. Thank you, Jason. There you go. It's right behind them.
Dr. Jason Barritt (17:25)
So there's a full micromanipulation
setup. You work in three dimensions at magnification, as well as you then control the fourth dimension, that's moving the fluid up and down the micromanipulation pipettes. But you have to be perfectly focused. You have to be able to handle what are here, multi-dimensional joysticks in each hand continuously while you're looking through a microscope at high magnification. And focal planes matter.
and then you use the thing on the very end for moving fluid up and down. So you're working in four dimensions on both hands simultaneously. It is a major thing for a human to learn this. It honestly takes a human usually about three to four years to actually be good at any manipulation. had to teach a computer to do this and then control it all. It's an absolutely amazing thing what's been done.
Dr. Jacques Cohen (18:10)
Yeah. Yeah.
Yeah, so good to have one behind you. I have some paintings behind me. You have the real thing behind you. So imagine that setup, Griffin. We had to automate all of it. So rather than hands, all of that had now to be connected with motors and microchips and computers and cameras instead of eyes. Every little opening there, the oculus, two cameras on their side port.
perhaps another camera there. So you're completely taking over the entire apparatus and get control over everything. You can move the stage automatically, can change the lenses automatically. We couldn't touch anything anymore, everything was automated in exactly a setup that looks like what Jason has behind him. Digital control, if you don't automate it, if you don't make it autonomous,
each little step, it becomes really difficult to have digital control. If you ever were involved in driving your first time you were in a Tesla and you were driving it, it's very difficult. Because you want the window washer, you want to your windows, wipe the windows, it's on a pad. So you're going to look at the pad and you get in an accident. Everything is on
So that's how that works, digitally control. We don't want the necessary digitally control in the way we were doing it at that point. So we're going to change that. But you asked me about what is the next step. The next step is going away from changing this existing system and building it from the ground up. And we call that Aura. And Aura is a line of systems. have Aura egg, which is doing the egg finding.
Later removing all the nursing cells around the egg, which are called cumulus cells, removing those, difficult process. Ambiologists have to get very experienced doing that well. It's very easy to damage the egg. And then sperm prep, which means you have to remove seminal plasma, which is the fluid, from the spermatozoa. And you have to try to take out the best spermatozoa and make those ready for ICSI.
People don't talk about it much. It's incredibly important. And that had to be automated. And then the XC process, which is 70 major commands, but I Jason alluded to it. He said 500, I agree with that. In code, it is at least 500. In terms of the tiniest step or small steps, it's about 100. In terms of actually clicking on...
So you can click on something that says immobilization, which means sperm, of course. You need to immobilize them before you do ICSI. And that also activates the sperm when you do that. And without that kind of activation, making it ready for fertilization, fertilization won't take place. So we had to automate that. And we had to automate finding the sperm, selecting the sperm. So for that, have AIs. We have an existing AI, fortunately.
which the program started with, called SIDS, sperm ID is what it stands for. Alejandro Chavez Barriola was the one who came up with that concept and it's great AI. You can use it by itself in an IVF lab, but in this case it was integrated. The best sperm is selected by the system without human interference and it's based on mortality, also on the way they're shaped.
based on motility, it's picked and it then has to drag it to the middle of the visual field while it is still motile. And in the middle of the visual field is a little laser that calculates where is the middle of the tail and lays a little bit of the middle of the tail very quickly, all of this in microseconds. So that's what that did. But we are now building that from the ground up. We're not anymore.
We're not anymore going into an existing lab and taking the existing equipment and changing that. Way too complicated because everybody has different equipment. So we're building this from the ground up. So you have the ICSI station, you have a fictification station, you have an egg finding and egg denudation station, you have a culture station and putting those in the line and behind there is a robot. One of our favorites because it's the only one that really is moving.
out of its station, it's called Handler. And Handler takes plates out of one station and moves them to another station. And the plates hold the petri dishes and where the culture is done or where the procedures are done. So yeah, that's what we are developing now, this ORAS system. And we're doing a second pilot study to make that work. And that's from the ground up and behind it all,
moving the apparatus around and doing everything. All these handling is called the Nexus system. It's a software system we're developing from the ground up with AIs being enforced in almost all of these stations, all of the stations. So very complicated orchestra to keep that going. It's tested, it's tedious work. You have to know if single systems work and you have to know whether in combinations they're.
So that's been going on for more than half a year and will continue for a bit.
Griffin Jones (23:43)
So
I want to talk about that next trial, but what Jacques is talking about with regard to no longer building the aura system in an existing IVF lab, but rather building a new, is that where we kind of left off in our last conversation, Jason? Were you talking about a hub and spoke model? that what you're visualizing with that?
Dr. Jason Barritt (24:04)
Yes, ⁓ the idea here is that, and this initial paper here is not that situation, it's what is coming from this initial paper, is the demonstration that when you can automate it and you can have it operate efficiently, you become no longer the limiting factor in where you can get care and what can happen. Because you can put it in different places.
and therefore you can bring the patient to it instead of having to build it everywhere. Because the truth is, this is not an easy task and it's not gonna take no time and it's gonna cost a lot of money. But the efficiency of it is what will make this a hub and spoke model. An aura system will sit in some major city or cities and around those cities potentially even.
And the vast majority of patients will come to it in order to get the care rather than an aura system being put in all 500 IVF labs across the country or thousands in the world. It won't be at all like that. It's just that the scalability is not possible. And you wouldn't be able to do enough patients in each center to actually make this an efficient, well-used thing. In this paper, I know it sounds really shocking. It's like 12 patients, 11 patients involved.
Dr. Jacques Cohen (25:04)
. you
Dr. Jason Barritt (25:24)
We're talking thousands that will be able to go through a machine in a year instead of 11, and type thing. And so the scale is huge, which will allow so much more access to care and will absolutely, ultimately reduce cost to do this. And the funny thing is, think Zock sort of mentioned it, it will take out the variability of us humans. Because truthfully, we're the ones who are quite variable.
And when you let a computer system and an automated system and then an AI controlled system, it actually can do better than us already, is what they've shown.
Griffin Jones (26:01)
So does this give you as lab directors more control? You feel like that sort of hub and spoke system? Because the way I perceive it right now, and maybe I'm wrong, but at least in the United States, it seems like the lab is the attachment to the clinic, not the other way around, that the owner of the fertility center is almost always the REI, and very often the lab director doesn't even own equity in the overall practice.
Dr. Jacques Cohen (26:23)
Yeah.
Griffin Jones (26:28)
And now
you're going to have the lab as the central point and different clinics plug-in. Does this give more control to lab directors to say, this is the way we do things. And then you figure out how you're going to do all the other stuff on the clinical end.
Dr. Jacques Cohen (26:46)
Yeah, if I may say something, I do want to give an observation on the how Harbin spoke. That model is in existence in the Netherlands since the 1980s. And there are several publications from the 80s, unfortunately not followed up.
People don't realize this, the entire country, it's a national health system. It's very different from the health care system here. But that is organized in 13 hubs. And all of those have spokes where the egg retrievals are done and the embryo transfers and the follicular stimulation. And those 13 hubs are the labs. There are rooms there to do also egg retrievals and to do also embryo transfer. But there are other hospitals.
other clinics that feed in the eggs and sometimes take the embryos back because they can be thought locally. So that is an existing system. So on the national level, that has been working there for them. Difficult to follow the national data. The advantage we have is the CDC data and the SAR data is just enormous. We know how well we're doing or how poorly we're doing, and therefore we strive to be the best.
not as good in other countries except for maybe some. So it's difficult to say if that affects the system, this harp and spoke model in that way, which is transport acts. You're transporting acts to the laboratory, but that can be done safely, can be done safely and the conditions can be well maintained. And so I think that's a country to look at in terms of harp and spoke.
So if you now add automation to this, I'm sure that the Dutch government, I think, will be interested in that. You add automation to this, you're driving the process to ultimately not to eight hours a day or 10 hours a day, but to 16, 20 hours a day. And in the case of setting up, preparing for the case, I left out what we call C-dish. C-dish is the conceivable way of...
or preparing the cases dishes. Tedious job. Ambient still like it. They don't like it. They run for the access when you say next week you're preparing the dishes each day. They run for the exit. So it's not something they enjoy. It's tedious. It's programmatic. It is ready for automation. So that you could do if you automate that, we are automating it. That you could do all day round.
It doesn't have to be in that window where everybody is saying, well, at the end of the day, I have to prepare the dishes. I'm tired. So it's often neglected. And it's incredibly important. Every step in the lab, every step in the clinic is incredibly important. You can't leave anything out. Everything plays a role. And so this star variation is determined by these many steps.
And with a system like Aura, we hope, on the supervision of embryologists, possibly remotely, right? You don't have to be there as a lab director. For Jason directing 19 labs, think, 19 labs, he can just have his iPad or telephone at hand wherever he goes. He probably does that already and uses the EMR.
But now you extend the EMR to something much bigger. You can actually control everything and you will be able to see everything. Because everything, you can't do a procedure when you do a procedure with humans, which is the standard, has worked very well. You see everything. So you can't have a black box and not see things. So you'll have to have cameras everywhere, at every position and get reports back so that somebody like Jason can do the entire country.
from the convenience wherever he wants to be, wherever he needs to be. That doesn't mean you exclude embryologists. You need embryologist local supervision, or you need embryologist with expertise like Jason's to supervise the entire thing and direct everything and ask the questions that are not maybe delivered. So I think ⁓ it's a completely different way of looking at the IVF process as it's done now.
There are examples of it like the Hop and Spoke you just discussed. There are examples of that around the world. We very focused into the United States, but we have to open up because it's done differently in other countries. In the case of the Netherlands, I think that is interesting what they have done since the 1980s.
Griffin Jones (31:15)
So that
and hub and spoke in the Netherlands might mean one thing because you can drive from one end of that country to the other in four hours, right? And I can't even get through halfway through New York state driving that long. And so, Jason, in this country, or at least in this continent, do you need to have.
that volume and scale in order to have a true hub and spoke model. Because we'll say, some people will say, we do a hub and spoke model. All they mean is like, you know, our lab is in Chicago and, you know, we have an office in Milwaukee and I'm not picking on anybody, that, or, you know, we have a lab in Boston and we have an office in New Hampshire. And that's what people say when they mean hub and spoke, but a real hub and spoke is that you,
you've got massive volume and then you have a system that allows people to use that as a reference lab from all different types of clinics. Is that right?
Dr. Jason Barritt (32:14)
Yeah, that's much more correct. So I'll give the example. I ran a center for 11 years here in Beverly Hills, California. We served eight, seven or eight internal physicians at that location. And we served 19 physicians who brought their cases to that location in that laboratory. That allowed them to stretch out, go much farther out and farther distance away from the lab. The patient really only, probably has to come two times to that location.
one for the retrieval and then thankfully when we put the embryos back in and they get pregnant, they don't ever have to come back to us again unless they're for their next one. So the idea is that, I'll call it around LA where it can take three hours to get three miles. The truth is you only have to do that once and you don't do that on a daily basis and you don't have to build all these labs all over LA just to serve it. It could be one location and you could serve, I'm not gonna truly say the number here, but you could probably serve
Dr. Jacques Cohen (32:59)
.
Dr. Jason Barritt (33:10)
half the population that are getting daily retrievals done in LA at one location. And as Jacques said, a clock matters, the hour of the day matters, but the truth is the machine goes 24
hours a day. It doesn't need a break. It's not gonna take lunch. It doesn't worry about its dog at home. There's nothing to this system that limits its capability of scaling. And that's what really, really allows Hubman's Boat to work even better.
Dr. Jacques Cohen (33:22)
Okay.
Dr. Jason Barritt (33:37)
I know it sounds terrible, but you can do things at six a.m. in the morning, a lot of surgeons do, by the way, and you can do things all the way at nine p.m. at night. That extends the day and how many patients we can see, how much stuff can be done, because the
Dr. Jacques Cohen (33:43)
Okay.
Dr. Jason Barritt (33:51)
machine doesn't get tired. It can go all the time. Therefore, you're able to serve stuff more.
Dr. Jacques Cohen (33:55)
Needs to be serviced, it needs to be
we're going into a direction of 20 hours a day instead of 12. And ultimately you go beyond that because you'll have twin systems. So one does these 30 hours and the other one does the other 20 hours or 15 hours each. So.
So you will have that, you'll have that. And I think that means you're servicing all around the clock. And that also means that people from other countries with experience can oversee it. You go 24 hours a day. Yeah, so Jason could labs in Europe at his leisure around the time when he is the middle of the day.
Dr. Jason Barritt (34:28)
⁓ there's the other thing is, yeah, I still have to sleep.
Dr. Jacques Cohen (34:39)
or the end of the day, five o'clock in the afternoon, he's doing labs in Paris. So I think that's the strength that I think will come out of this plus the standardization. Those are two big things. will ultimately, Jason used the word ultimately for driving down the price, driving down the cost. And I think he's right about that.
Dr. Jason Barritt (35:01)
Yeah, it's gonna take some time on that one. The other big factor here is where you're going Griffin, I think is actually what I'll call the large fertility networks, especially in the US at the moment. I don't know the rest of the world as well, but I mean, EV is a big one, of course, but those have to work on scale. They have to have many, I'll call it feeding clinics into the main place in order to be the most efficient.
That is where all the networks will want to go. They will have their main hub in LA, Chicago, New York. know, there'll be the main hubs there and everybody will come to them in that network. The networks will absolutely want to do this because they can scale up so dramatically and help so many more patients. It might even allow multiple networks to come to the same one, which would be even more cost effective to be completely honest. The truth is everybody building a car by themselves
Great, beautiful. But the truth is, we don't use our car probably 95 % of the time. It has been assembled and is sitting there. Therefore, the efficiency is horrible. We need to find a way to make it more efficient and bring the patients to the unit, which will allow it to happen.
Griffin Jones (36:14)
a drawback to the embryologists having more control, like being in the driver's seat of the lab being the reference lab, of them being the hub and the clinics being the spoke as opposed to having more of a kind of one-to-one relationship that they've more or less had in the current fertility center dynamic.
Dr. Jacques Cohen (36:31)
Well, I think
you described, it's interesting what you describe and how you're formulating this. I think that situation already exists to some extent. Doctors are very dependent on embryologists. They are looking and keeping them happy, trying to keep them happy, trying to keep them interested. So I think that is not really going to change going forward. I think
I think we'll move automation out into the clinical area. That's already happening with some AI, quite a few AIs that you see in place or available for doctors to do, instance, help them with follicular stimulation, standardized follicular stimulation, because often done in a way where one doctor on duty one day does something else than the doctor the next day. And it's a complicated thing, actually, follicular stimulation. It's not really a truly
100 % standard operating procedure. So, there are AIs helping with that. There could be AIs helping with the accurate retrieval and semi-automated processes with the accurate retrieval. I think it's going to take some courage to automate an accurate retrieval. I'm not saying it's not possible. It is possible, but it will take courage and it will be difficult from a regulatory point of view because...
If you look at the history of the Avinci, regulatory is in charge there. And it makes sense from a liability perspective, it makes sense for the safety of the patient. And so it's going to be a bit more difficult to automate or include, introduce automation on the clinical side, but that is going to happen. I don't think that when you get automation in the lab, that that means the embryologists are in a better position. They're probably happy to hear that from you, Griffin.
We're not listening to this broadcast, but I think that we will have to wait and see. They are actually, a lot of embryologists, I think are somewhat afraid of automation, not only because they think they are going to be replaced, which is absolutely not the case. Their job is going to change. That is scary enough. Once your job changes, you have now, you have learned this for five or 10 years, this is what you're doing, and now that is going to change.
That is involving, they become more like engineers. So ⁓ it's not just embryology anymore, they become more like engineers. We call them embryoneers, by the way. It's a terminology that we are using for those people in the future. And I think the job of embryologist is going to be a lot more interesting. This is the best time to become an embryologist. There's no doubt about it. This is the best time.
and it has improved in that quality in the last 10, 20 years. But there's also a lot of stress and we need to take that away. I think automation in part will take that stress away. So you don't have to do the sperm preparations that you don't like to do. You don't have to set up the dishes. And at some point this goes well beyond conceivable and other automation companies. But at some point we hope to get to a point where we don't have to look at the monitor 50 % of the day.
because that's what embryologists do. They look at the monitor 50 % of the day, putting in data, doing quality control studies, ordering stuff. All of that needs to be automated. They want to get rid of that because that's what's stressing them out. It's at least 50 % of the day. Doctors spend 50 % of the time in profession behind the monitor. That's why when you go and visit a doctor, they're often looking at the monitor and you're sitting right behind them. Next to the monitor,
And you get a glance or so. That needs to change. The things that we need to type in all the time, we have this incredible AI that is skyrocketing right now. And we're still typing. We're still using a mouse. That's slowing us down. That needs to be automated. I think once you do that, and that's going to be done, I think in the next 10 years, five years, this is starting to happening already.
People are thinking about it, technologies are thinking about it, using that. The job of embryology is only going to get better. Of course, less monitor time, more action, more supervision, more intelligence. It'll really good fun.
Dr. Jason Barritt (40:37)
Yes,
they are going to advance. The human is going to advance in this also. We are going to be engineers, reproductive engineers, and we're going to help make it better. So the big thing about any of the automations and any of the AI is we're going to take out variability. It's something we monitor every single day. We spend probably an hour at every single location doing quality control to make sure all the equipment's functional before we even do anything.
This thing can do it all the time. It can monitor everything all the time and it doesn't take a human to have to do it. And it can be adapting to something that might not be working much faster than we can. Therefore, it actually has efficiency and scale. all the way back to your question is, the embryologist is actually gonna love this. But yes, it is gonna be a change in their careers and what they do as a hands-on, daily, everyday job.
but it's gonna bring so much consistency. So back to where your question was, the embryologist is per se gonna be in control of the fact that they have this hub location that everybody wants to come to because that's gonna be the best place they have the option to bring anything to anyway. It's gonna be so consistent and so reliable, it takes out all the variables. Therefore, they get to be the best doctor they can be and bring it to the best place it can be done embryologic.
Griffin Jones (41:56)
What needs to happen next in with regard to the trial? So, Jacque, it sounds like you're working on a trial and I know that people are kind of sometimes restricted in what they're able to say and in what they're working on. But again, this is me coming from someone that doesn't know how these types of studies work. Do you just repeat the same trial with a larger sample size or what will be different?
Dr. Jacques Cohen (42:20)
Yeah. Yeah.
Yeah, well, it's definitely that. It has to be a larger sample size. When we did ⁓ work with HOPE in Guadalajara, the HOPE clinic in Guadalajara.
They love minimal stimulation. They have great results with it. It's a rare approach in the United States. actually, one thing Jason said to me not too long ago, what is amazing about the paper is that it involves so few accidents because we do minimal stimulation. Incredibly challenging. So we're moving away from that in the other whole clinic in Mexico City where the first oral line is installed and it involves dish preparation.
It involves sperm prep, egg prep, egg denudation, ichthy, and culture, and vitification. It's a lot. So you have to have lots of eggs. If you ask me what is my biggest challenge, oocytes, my biggest number of oocytes, that's my biggest challenge. So this requires a lot of observations, and we're developing the technology. And in the meantime, you don't want to jeopardize the patient.
So you have to make sure she gets enough oocytes from the manual conventional IVF laboratory there that does really, really well. And so that of that is going to play a role in the next year in this trial or the next few months in this trial. It's a process of development. Is that usual in trials? No.
trials is you actually, you do randomized trials where you have a technology that several people have published about and you say, well, but nobody has really done a good randomized trial. So let me do a randomized trial where I have an arm of patients that getting the conventional treatment and an arm of patients that have that one thing added to the conventional treatment. And then when those patients are comparable, then I can compare them or you do a trial, which is the same as in this trial.
where you have what they call a sibling oocyte study, where you take half the oocytes from a patient and do your new work, and the other half of the oocytes you do what you normally do, the conventional regular approach. And so that is in place. We take acts for the patient through the regular IVF lab and compare them with this very advanced new system while developing the technology and learning from the first patients we know more.
for the next 10 or the next 50. So yeah, this is a process. Clearly more complicated than the first study, which we are making use of existing equipment that for an IVF lab works. And we're making use of that, automating that. Now we're building it from the ground up.
Griffin Jones (45:02)
The last study took about six months, the RCT itself, and then it was another year and a half, give or take, before it was published. Should we expect the same timeline for this one? Could it be even longer?
Dr. Jacques Cohen (45:13)
Yeah,
it will probably be a little longer. Yeah, it's like that. But we hope to take the second version of what we have now into the first lab in the United States, and hopefully not long after the second one, at least a good portion of what we have, and then take that, install it, and then those centers, I think, no doubt will require or demand
that they do their own trial. And so you're looking at a process like that. Also, I want to add that if you have automation done at some point where you say, well, I have my platform, now I'm done, can improve it, optimize it and improve it. The advantage of that is that anything new that the field produces, like a new culture medium that people have tested in the mouse and are crazy about.
Well, if we want to do that in human, automation and automated laboratory is the place to test it because you have removed a lot of factors. A lot of these variables are now gone because that's the problem when we do randomized trials. Good randomized trials always involve more than 500 patients. And the reason is that there so many variables that if you look at a population of just 20 patients, you're not telling anybody something that's new.
you're looking at a bag full of variables and we think there are hundreds of them. And so you need to do a lot of patients to do a really good randomized clinical trial for one, for testing one item, whether that's a hormone or a drug in a clinical lab or in a clinical environment, or whether that is a single step, something new in the embryology environment. So I think automation in that respect will make a difference, that and the worldwide system that it can unfold.
and the 20 or 24 hours a day that these labs can operate.
Griffin Jones (47:05)
Would it be accurate to summarize your point, Jacques, that an automated lab would be the supreme testing ground for new point solutions in the lab?
Dr. Jacques Cohen (47:14)
Yeah, I hate the word point solution. You must have noticed I'm talking around it. Point solution sounds always like something little, but behind on the right of Jason is standing what you could call a point solution, right? It's just that blue box, that beautiful blue box with the smile, the ambioscope. That's a point solution. So let's not underestimate what a point solution could mean. When we start, when we went from
the precursor micro-replacement systems in the 1980s to ICSI early in 1992. Those were all point solutions and ICSI is a point solution, it does take over the world. So I don't like that combination of point and solution. doesn't mean, okay, clear, the aura is not a point solution. That's maybe 50 point solutions or 40 point solutions, but...
I have never been a fable. Yeah, yeah, okay, okay. Yeah, yeah, okay.
Griffin Jones (48:07)
So I'll get rid of one of those words and we could say that an automated lab would be the supreme
testing ground for either points or solutions.
Dr. Jacques Cohen (48:15)
Correct, yes, thank you.
Griffin Jones (48:17)
Jason, what threshold would you want to see from from the next study to say that if they do X
that would mean that it's ready for prime time and if they don't do X then it's not ready yet. What threshold would you set for them?
Dr. Jason Barritt (48:29)
you
they want to be non-inferior to us, the human. That's where we're gonna start. Well, let me just highlight a couple points that are in this paper. Remember, I'm not an author. I didn't do this work. But I look at it and I evaluate it and I figure out where it's gonna happen. So first off, safety. They did not damage a single oocyte in the full process of denuding.
the eggs. I hate to admit this, but humans sometimes damage things. Not one. Hundreds of eggs, hundreds and hundreds of eggs didn't damage one. That is a huge leap and that already tells me we're succeeding. They had five healthy births already. That's telling me that everything they made from the second they caught a cumulus complex all the way through the baby worked.
Dr. Jacques Cohen (49:00)
Thank
Dr. Jason Barritt (49:23)
Every step worked because any failure, any point failure, would have resulted in none of these, which means they've worked out so many of them. Technically, I'm gonna come back to the, I always think we can all do better. I mean, the automated fertilization rate, now remember, they caught the eggs, denuded the eggs, they processed the sperm, they got the sperm, they automated the Ixie, they selected the one, got it all done, and then cultured the things on.
they fertilize 64.3 % of the eggs.
All of us are shooting for higher numbers. Of course we are. But it's well within the Vienna consensus for an expected fertilization rate. So the truth is they may have already got us. The machine has already got us. And then let's talk about their usable blast rate, which is a hugely important thing for all of us. They have 42.2 % usable blast rate. That is a hugely positive number. Is it?
Dr. Jacques Cohen (50:03)
Yeah.
Dr. Jason Barritt (50:18)
still lower than we're all shooting for every day. Yes, and there are clinics that are achieving a lot more than that, who are unbelievably controlled and have had years and years and years and years, 30 years of work to be that good. Their machines there now. Imagine where they're gonna be in three years. Or wait a second, that three years has already passed, they're already there now. That's why I'm saying that the paper helped me get to, this is the learning curve.
This is teaching a kid to walk, but now they run. So guess what? We all learned to learn to walk. We all got there at different time points. This thing's already there. It's had three years. It's already running like an Olympian and achieving Vienna consensus numbers of an average human already. Imagine that it will win the gold medal when they have enough time to make this thing work the right way and have 5,000 point solutions.
Because that's truly what this is. This is all of them at once in order to improve overall outcome. So look, I look at this as a, a huge proof of concept, but I just got to watch it learning to walk and it can walk already. Is it gonna be that some, richest man in the world or the richest person in the world can decide to...
fund a whole bunch of robots and make all this happen in a way. Sure, that type of thing could happen, but that's probably not the way this is all going to go. This is going to go through a lot of work by a lot of people, 50 to 100 plus engineers, all trying to figure these things out. And then automated systems that have been developed, AI systems that have been developed by other people, but then applied to what we do. All of that is huge amount of human power, a huge amount of computing power that is going to push us way beyond where we are now.
and make us better, more consistent. And truthfully, almost everybody's gonna want this. This is going to be the elite of it. If it's already walking, it tells me where we can go. And with the speed that AI is developing in six months or even a year, or even two years ago when I first played with ChatGPT, it is a completely different person now. It is better than me. It can come up with things faster than-
It knows more than me. It will know more than me ever, ever, ever. And it can learn everything that's coming out from every single paper on this field in one minute. And it can apply things and combine things that would take me my entire career to do. It can do it now. Now imagine putting all of that power into this. It can adapt to every bit of variability instantly. And it can know that it saw it hundred other times and knows what to do. That's where we're getting.
Dr. Jacques Cohen (52:31)
Yeah.
Dr. Jason Barritt (53:00)
with this type of system.
Griffin Jones (53:02)
And if Elon Musk is listening, maybe he'll decide that he wants to take advantage of Conceivables next funding round. That's right.
Dr. Jacques Cohen (53:06)
Write a check. Write a check. Yeah,
Dr. Jason Barritt (53:10)
I wasn't calling out any specific, by the way. Just the in general, thing is a humanity thing. Let's be completely honest. Reproduction is basically a human right. And we want to actually allow its access and allow everybody to have that access to it. The truth is we're trying to overcome where biology is limiting this.
and we're finding ways to do that. And the truth is, we're all having to work very hard to do that, but we already succeeded. Jacques was there at the beginning of IVF. Imagine where we are now. They weren't not thinking of this level and this many millions of babies and everything else at that time. Well, now imagine we're sitting right now and we're thinking about, well, it's a million babies a year. We're gonna talk 10 million. We're gonna talk 50 million because this will be able to do it and it'll be able to do it safely.
accurately and in a way that's better than everything that we can do it right now.
Dr. Jacques Cohen (54:04)
We had to learn artificial insemination, assisted reproduction after that, and are finding, using the least fertile patients, are finding that the results are due to nature and because of
Dr. Jason Barritt (54:18)
Hmm
Dr. Jacques Cohen (54:24)
The testing that's done and all the diagnostic tests that are being developed, we think that in due course, this will be considered the safest ways to reproduce. I think that is what Jason is saying because he's going from 1 million to 50 million. That is not covering infertility. He's saying, why would you do anything else? We're not there yet because I don't think you want to come to a clinic 10 or 20 times or five times.
You really want to donate probably some cells for both male and female partners. So we still have to go a ways. We still have about 20 or 30 years before that's going to happen, I think, and maybe longer. But this is going to be the safest way to reproduce.
Griffin Jones (55:07)
Dr. Cohen, Dr. Barritt, I've had you on before and I'll have you each on again. Thank you so much for rejoining me on the Inside Reproductive Health podcast.
