BOV 69 | AI For Fertility

 

What do automotive technology and fertility management have in common? As unlikely as this pairing sounds, they both benefit from the power of artificial intelligence. This is why it was quite seamless for Eran Eshed to shift careers from the technology space to fertility management. Ultimately, the fundamentals of the technology were the same, and Eran was just happy to be able to apply his expertise to something that’s very compassionate. Joining Alyssa Dweck and Rachel Braun Scherl on the podcast, the co-founder and CEO of Fairtility shares how his team in Israel developed a way to enhance the process of creating families with the use of AI. He introduces the platform they’re working on that benefits three main stakeholders: embryology labs, physicians, and aspiring parents. He also touches on some of the more controversial aspects of AI in fertility like genetic testing and the impact of AI on the jobs of physicians. Tune in to take a peek into the future of fertility management that’s already happening now!

Listen to the podcast here

 

Fairtility: Using AI For Fertility Management With Eran Eshed

New Approaches In Family-Building

Rachel, in practice over the years, we always categorize fertility or infertility into three categories. We would say there could be a female factor, whether it’s a lack of ovulation, a hormone imbalance, or something structural going on. We talked about the male factor with difficulties, with sperm quality, quantity, and the like.

There was the big black box. One-third of couples having trouble getting pregnant turned out to be undiagnosed and unexplained. This is incredibly frustrating and emotionally frustrating for my patients for sure. When we speak to Eran, we are going to learn just how far we’ve come in a way to assist particularly in that unexplained category to help couples or individuals with their fertility.

One of the things that I found along those lines so interesting is we talk about in so many of these situations and health challenges that women face that are so multifaceted. There are emotional, financial, and physical aspects, but what’s interesting, talking to Eran who founded his company in Israel, is there’s still emotional and physical, but there is not the same financial burden because the State of Israel pays for quite a number of cycles.

 

BOV 69 | AI For Fertility

 

It’s interesting to keep that in mind, depending on where you are in the world. The level of pain of certain of those factors could be different. This was like a masterclass in Genetics, which I clearly could have used some more time in in college. He’s really deep in science. He’s careful about what you can say and you can’t say, because you don’t want to make any scientific claims that are not validated. It’s an interesting way to see where science could ultimately be taking us.

One last thing to bring up that I’m happy about is that AI is going to help clinicians with managing fertility for people, but not replace them. I don’t think we’ll be out of a job anytime soon.

That is interesting because that’s a theme that comes up a lot. With so many of these direct-to-consumer or digital health, we want to put a lot of the power in the hands of the patient. At least in the context of our conversations and the work that you do day in and day out, that is never meant get rid of the physician altogether. The physician still plays an important role in so many of these areas.

 

BOV 69 | AI For Fertility

 

We are so thrilled to welcome our guest, Eran Eshed, who is the Cofounder and CEO of Fairtility. We’re excited. Thank you so much for being here.

The excitement is mine. Thank you for having me.

Let’s jump right in and talk about Fairtility, how you got there, how you decided this was a need, and what it is. For both of us, Alyssa and I, there is a lot of exciting activity in this space. We’re interested to know what piece of the land that you’re particularly focused on.

Like many things in life, this was not intended. It happened by chance that I met my two cofounders, one of whom is a fertility specialist. He is a doctor and the Head of the Hadassah Fertility Center here. It is one of the largest and best ones. Another one is the CTO who was doing his PhD work on embryo assessment. I’ve been in the technology space for over 25 years. I’ve done many things from semiconductors to embedded software. I even did artificial intelligence in automotive.

When I met them by chance, I was blown away by the prospect of how we could apply this to something that is better and more compassionate than eCommerce, FinTech, or crypto. They all use, in many cases, the same technology fundamentals. In this case, the idea of how we could help a couple of women to become parents with something like AI was flabbergasting to me.

 

BOV 70 | Menstrual Cycle

 

We often laugh when we hear people say, “How’d you go from Microsoft to menopause? How do you go from automotive to AI for fertility?” It’s great to have you. I’m going to turn it over to Alyssa. You see fewer patients at this point in the practice who are trying to get pregnant because the majority of your patient population is older. How frustrating has it been as a practitioner to not have more solutions? We’ve made a lot of progress. I don’t mean to suggest that we haven’t. How does this fit into the conversations that you found yourself having with patients?

As a matter of fact, this is top of mind for so many of my patients who either are pre-pregnancy or maybe have been focusing on their professional lives and they want to freeze their eggs. The conversation for that type of assistance in fertility starts as young as age 30 or 32, which was unheard of when I started in practice. That is something that’s top of mind for so many of my younger end patients. I’m curious if that’s a similar conversation that’s going on in Israel or if it differs from the US. If so, is cryopreservation part of your assessment and your expertise in the company?

First of all, the answer to pretty much everything you asked is yes. The differences between the Israeli fertility market and everywhere else, specifically the US, are a conversation for a full episode.

I’m sorry about that.

It’s quite interesting. Israel is an anomaly. In this case, for various reasons, treatments are 100% subsidized by the government, which has its own objectives. I don’t remember the number, but you could have up to six children and you’re allowed to continue until you have children of both sexes. It’s quite amazing. When you compare this to the US, it’s a completely different opera. On a global scale, our services are provided in many countries. It’s soon in the US, but we’re in many countries in Europe, Asia Pacific, and the Middle East. We’re exposed to a lot of data. We see the age distributions. We see many parameters that pertain to being able to understand the dynamics behind what is happening, why people are doing this, what ages, and so on.

There’s a trend towards becoming parents later. There are many different reasons for that, which lends itself to a lot of egg freezing. It happens in ages that are even younger than what you mentioned, which is 20 or mid-20s. We try to address many of the unmet needs in these markets. This has nothing to do with fertility or with Fairtility. It has to do with a general concept of when you decide that you want to invest your time, money, and investors’ money in a certain endeavor, it is when you check three boxes. You have to check all of them.

There is a trend towards becoming parents later, which lends itself to a lot of egg freezing. Click To Tweet

1) There’s a real unmet need. 2) You’re able to do something about it, mitigate it, solve it, and improve the current standard of care, whatever that may be. 3) You can make money. There is unit economics and a business model behind it. In this case, there are so many unmet needs that it makes it even more difficult to decide where you want to focus.

I love this mission and the way you organized it in those three steps. I’m thinking that the younger people who have an indication to have a need for this would be those with some hormonal issue, perhaps polycystic ovarian syndrome. They are those, from a women’s standpoint or women’s health standpoint, dealing with age-related issues, fertility, and then male issues. Can you take us briefly through the journey of the person who wants to get involved as a patient? Is that something you can take us through?

First of all, it does vary from country to country the way people choose their journey, the way they select which physician to turn to, and the way they’re referred through that, whether it’s through searching the web, word of mouth, or they were referred to by their OB-GYN. Once they do choose what clinic and what physician they want to go to after having established that there is a problem, then they start a journey I’m sure you both are very familiar with.

It’s a long journey. It’s complex. It’s expensive in most countries. It takes a huge emotional and physiological toll. It involves many different touchpoints for the patient with a clinic, a physician, and a lab in which there’s little transparency. That’s one of the issues. When we mention an unmet need, it’s not about how we improve the decision-making process, which is something I’m sure we’ll talk about in a moment. There are many decision-making junctures along the journey, which are not done in an effective way. Even in the engagement level, that is the transparency that patients have.

They would start with the consultation in which they have been told, “You’re this age. You have that BMI and this AMH and AFC levels. Let’s try this protocol. Take this drug. Let’s give it a try, see what happens, and adjust as we go.” We know how poorly that works because sometimes, you understimulate, and sometimes, you overstimulate.

There’s so much research around does a higher dose of drugs produces more eggs, or are more eggs better than maybe slightly fewer eggs but of higher quality. You start with that. There’s the retrieval in the fertilization and the embryo culture, which is one of the areas where people don’t know what’s going on there in that back room. Sometimes, they would get a call from the nurse practitioner or from the lab staff. They tell them something that they have to google and try to figure out what that means. The healthcare professionals themselves and the embryologists don’t know. They have rudimentary ways of assessing quality and potential. They don’t know much, and patients know much less.

At the end of that 5 or 7-day period, some selection is made, whether it is a fresh transfer, a frozen transfer, or whatever the case may be, but you’re in the dark. You’re hoping. Your level of anxiety is very high. Once something fails, there’s little explanation that you’re given on why because they simply don’t know. This is where we get in. The mission is, first and foremost, to improve the process and the decision-making and make it data-based and data science-based. We apply advanced AI techniques to do that, which translates into a shorter time to pregnancy, hopefully, and improved rates, but not less important. It is an experience for patients that want to continue trying.

Fairtility’s mission is first and foremost to improve the fertility management process and improve decision-making by making it data science based. Click To Tweet

Let’s try to break it down for people who don’t understand the process. Those are the people who are in the space as investors or entrepreneurs, and importantly, patients and physicians. How does it work? How do you apply AI? What does the patient do that’s different? What does the physician do that’s different?

It’s not a single answer. When you look at the IVF journey, there are arguably 5, 6, or 7 different paradigms in the process that can be automated and powered by data and data science. In each of them, it means something different. I’ll tell you a little about what we have and where we are planning to go. Where we are is with an embryology assistant called CHLOE. CHLOE is installed in embryology labs. It connects to different sources of data in that lab and, most notably, incubators that have optics in them.

I was interested in reading about it. What does the acronym stand for?

I’m embarrassed to say that I never remember.

That’s okay.

It’s because we retrofitted the definition. This is part of when I said engagement and transparency, it’s also a personal touch. This is not an AI algorithm telling you something. It’s an embryology assistant. It’s like Siri in a way. Embryologists interact with it. Patients interact with it. It’s like a human embryologist giving you information and insights. CHLOE analyzes images, videos, and different clinical parameters from the EMR and EHR systems. It also provides embryologists with an ongoing assessment of the quality of the embryo or the developing embryo throughout the process.

I have one question for you. I’m not a reproductive endocrinologist. I don’t do IVF, but have a lot of patients who have undergone this. Where does checking for genetics come into this process? I understand that CHLOE may help with egg selection quality using traditional lab parameters for patient choice and whatnot.

Those are AMH or Anti-Müllerian Hormone, which is a typical test we do to check for the quality of eggs at the moment, and BMI or Body Mass Index because we know that people who have obesity may have a lower fertility rate. Where does PGD or Pre-Genetic Diagnosis get replaced by AI? Does it check for any genetic issues? Is it sheerly CHLOE that’s used to select the best eggs or embryos for successful pregnancy regardless?

It is applicable to the process that happens in the US because the US is intense on genetic testing. Let’s start by saying that genetic testing is controversial. It’s not a method that you would find broad consensus that improves the selection process. In many ways, it’s the only way that exists to validate viability beyond just looking at the image and saying, “This looks like a pretty good embryo. It’s symmetrical.”

Genetic testing is controversial. But in many ways, it's the only way that exists today to validate viability beyond just looking at the image of the embryo. Click To Tweet

For the layman out there to distinguish, there’s a big difference between looking for genetic mutations and designing a baby that has particular gender, hair color, eye color, and characteristics. Is that an accurate way to frame it? Lots of people are already doing pre-screening for the health of their pregnancies and the viability of their pregnancies. That genetic testing could be a whole other episode, but there’s some nuance there.

There’s a big nuance. It’s different, but it also overlaps a lot. When you look at genetic testing, the process is about biopsying a few cells from the trophectoderm of the embryo, i.e. the cells that will become the placenta and not the actual embryo cells. They’re sequencing them and trying to find chromosomal abnormalities.

The reason the test is being criticized aside from the fact that it’s expensive is that, first of all, you’re not biopsying the embryo. You’re biopsying the placenta. We all know that genetic sequencing is a noisy test. When you amplify DNA, you get misreads from a noisy amplification. The third thing is that it’s clear that being euploid, i.e. having a symmetrical set of chromosomes. Twenty-three chromosomes are necessary, but it doesn’t mean that it’s sufficient.

Finally, it’s researched, documented, and published that embryos know how to correct some of these abnormalities. It’s looking through a narrow lens at something and saying, “This is the only way we know how to check something that is indicative of quality. Let’s apply these criteria across the board.” There are many false negatives and many false positives. The accuracy of PGTA, in this case, or genetic sequencing for screening embryos has an accuracy of roughly high 60% or maybe 70%. Clearly, it’s not enough. There’s more out there.

The question is, what can you do more or combine? What we say, and this is something that we invest a lot of effort and energy in providing clinical validation, is that at the end of the day, this is about fusing different sources of data in different inputs. There’s not one single technique that can tell you what you need with 100% accuracy. There is no 100%, but we know 70% is not good enough. Can we take this to 75%, 80%, or 90%? This is what CHLOE does, which is based on morphology and morphokinetic analysis, i.e. images. The development of the embryo over time and applying this through AI models adds another layer of information that, in many cases, addresses these inaccuracies and improve the results.

I want to get to the business side of things, but I have one last medical type of question. Here in the US, reproductive endocrinology clinics or fertility clinics usually rely on publishing their pregnancy rates from IVF. Does CHLOE have a rate yet? Is there anything that is comparable, or is this not used in your neck of the woods?

First of all, there is SART reporting. In Europe, there’s HFEA. In the UK and other places, there are different systems. This is a problematic outlet because there are a lot of ways to count and report. It’s a problematic measure. It’s the best you have, but it’s problematic. If you want to measure something for efficacy, it requires prospective testing and usage of the system. The best thing would be a controlled clinical trial, but these things take years, especially in the case of fertility. If you’re looking for a live birth rate, you have to wait nine months. By looking at the inclusion and exclusion criteria, you’re looking at 3 or 4 years’ worth of a trial to get something that is statistically significant.

The real test is how well did you do before CHLOE, and how well you do after CHLOE, which may not be necessarily a very rigorous clinical measure of saying, “Here’s the difference statistically,” but it is a measure of the quality of outputs. We launched CHLOE commercially a few months ago at ESHRE in 2022. Since then, we signed up a few tens of customers who are using CHLOE in clinical practice. We anticipate that within a year or so, we will have these clinics being able to report, “This is what happened before. This is what happened after.” That data does not exist for CHLOE or for any other system out there. These things take time.

In terms of where CHLOE is in the marketplace, help us understand where it’s in practice and how it’s being used. Is it a specific number of clinics? Are you able to increase the capacity of the number of patients who can go through it? People who are struggling with infertility are often rabid researchers. We’d love to share some information about where this is. Is this an option? Do they have to fly to Israel? Do they have to be in the Middle East? How could someone find out about adding what sounds like a layer of assurance or increased probability of success to their fertility journey?

Being a software or medical device company, we are very prudent with saying, “We’re going to improve your rates.” You don’t say these things. There are things you can and cannot say. You won’t find us saying, “CHLOE’s going to improve your rates.” What CHLOE does has a value proposition to three main stakeholders. I’ll get into where you can find it and what it means for you, the patient.

The first customer for CHLOE is the embryology lab. You’re looking at embryologists that spend, and we’ve done a lot of research on that area, up to 48% of their day-to-day job in administrative work. It’s mundane, specific administrative work that can be completely automated. There’s no reason for a human to do this. They become a Tesla autopilot. They sit behind the wheel. They supervise. They don’t have to be driving the car. Everything is automated. They supervise it. It saves a lot of time and improves attrition. It’s a tough job for embryologists.

The first customer for CHLOE™ is the embryology lab. Embryologists spend up to 48% of their day-to-day job in administrative work that can be completely automated. Click To Tweet

You’re able to achieve a lot more in the embryology lab. In research that we’ve done based on clinics that use CHLOE, we see up to a 30% reduction in embryology hours per cycle. We see an increase of up to 50% in lab capacity. Let’s even say you have the data. You’re typing it into the EMR system. They write stuff on paper. I was shocked to see that. Information is written on paper and then typed into the EMR. Many mistakes happen.

CHLOE is always fresh. She’s not tired. She’s not annoyed. She’ll always give you the same answer because it’s based on the same analysis of the data. Embryology above the decision-making and efficiency, CHLOE provides a layer of BI or Business Intelligence. If you’re a lab director, from your smartphone at home, you can see everything that happens with the different benches and workstations in your clinic.

If you’re a network and you’re a scientific director or a COO of a network and you have twenty clinics, you can see all of them on a single dashboard. You understand what works better or worse and get real-time alerts if something happens. You don’t react a month after when you see fert rates drop. You get a real-time alert so you can address it. Everything that has to do in the embryology lab.

The second thing is the physician. Alyssa, although you’re not an REI, I’m sure you can relate to this. One of the pain points of physicians is not being able to have an informative conversation with patients after the cycle has not been successful. Unfortunately, we know the statistics. It’s about 2.7% cycles in the US to get a baby in case it’s successful. You’re failing cycle by cycle. Why?

It’s being able to have that conversation and say, “We did transfer a genetically viable embryo. It was not successful, but let me show you how that embryo developed and give you some insights. It developed a little slower than what we expected.” They don’t know these things. They’re avid, but they’re not reproductive experts. At least they get some explanation that they can go afterward and breed. It does reduce the level of anxiety. It helps physicians manage that conversation, which is one of these unmet needs that we mapped out.

The third one is the patient. The patient, at any given moment through an app on a smartphone, logs in, checks her embryos, sees the assessments and gets educational material. She doesn’t have to wait for a text message or a phone call on day three. It’s always there. It gives you some peace of mind. This is not scientific, but it makes sense that when you’re less anxious, you are more likely to be successful in such a treatment. All these together are the benefits that CHLOE brings.

If you want to use CHLOE, you have to be performing your treatment in a clinic that has CHLOE. It’s mostly in Europe. We have a CE mark that allows us to sell CHLOE commercially. In fact, you’ll see already in some countries in Europe that the clinics themselves launch marketing campaigns that position themselves as leaders in the market because they’re using cutting-edge technology or AI and it’s powered by CHLOE. It’s their own brand powered by CHLOE with all the tools that they use and can make accessible to patients.

Let me jump in there for a second. When someone sees that powered by CHLOE, and I know there are things you will be able to say in the future that you’re not able to say here, like it increases the rate and reduces pre-term labor, and I’m making these up to anybody who’s reading, you’d be able to measure them. What does it mean? What are you hoping it communicates when a clinic says, “Powered by CHLOE.” What is the communication to the patient? Is it implicit or explicit? If you are going to a clinic that’s powered by CHLOE, you can expect fill-in-the-blank.

It very much depends. Like always, different clinics have different ways to position themselves in the value and benefit that CHLOE could bring to patients. We work with them quite closely to tailor these messages under MLR. Those are Marketing Legal Regulatory frameworks that tell us exactly what we can and can’t say. If I have to dumb this down to one statement, and Alyssa I’m sure can relate to this, the physician or the doctor used to be god. They’re like, “The physician said so.”

There’s a generational shift. That doesn’t happen anymore. People read. People research. They want information and answers, from their perspective, when the physician or the doctor says, “We are equipping ourselves with the best technology out there to make your journey as successful as possible and as engaging and transparent as possible. Research shows that X, Y, and Z.”

If you look at the fertility space, it is regulated, but you would be surprised because I was, when you read, for example, the small print in the genetic test results. You’re paying $2,500 to get your embryos genetically tested. You get the results, which half of them say, “These are inconclusive. We don’t know.” At the bottom, it says, “It means nothing. We don’t guarantee anything. Some research says this. Some research says that.”

At the end of the day, you’re looking to gain as much assurance. You want to try as many tests and procedures to make sure you maximize your chances of success. For the patient, it means the clinic or the physician that you trust or that you want to trust takes their profession seriously and uses everything. They’re using the best technology and the most advanced researched medical tools to try and make your treatment successful and more engaging. That alone, as a statement, creates trust. You don’t want to put your car in a shop that doesn’t use computerized tools to talk to your car. Why would it be any different?

That’s a good analogy. As we always say, healthy mom, healthy baby. We have to figure out the best path to get there. You mentioned something about the third step of your journey and your business in terms of monetization. I am unfamiliar with having the state pay for all medical care. That’s a foreign piece of information to me here in the US. How does your company monetize it? Is it all through state-funded funds, or is there any private pay at all?

Unfortunately, fertility in the US is mostly out of pocket. There’s very little insurance coverage. There’s more of a trend for employer benefits that would cover some, but it’s out of pocket. The spending decisions are made by the patients themselves without having to worry whether it’s covered or not, because it’s not, unfortunately.

In our monetization model, we charge clinics on a per-cycle basis. They decide how they want to market and charge. Some would charge patients for an add-on. Some would bake it into the base price and say, “All cycles in our clinic. We use the best technology out there. If you want to get your treatment in our clinic, then there’s a payment.” Some would say, “It’s our own tool of being able to improve the standard of care. We’re not going to charge for it,” and they treat it as an expense. Either way, we get paid on a per-cycle basis.

What I mentioned we do is one first step in developing a much more comprehensive platform that addresses egg quality, sperm quality, automated stimulation, protocol generation, and best transfer timing determination based on uterus lining. All of these things, at the end of the day, are things that are practices that have been the same over the years. It’s all based on physicians’ and biologists’ experience. It’s qualitative and not quantitative. It is not standardized.

Establishing this golden standard of care on data, data insights, and digitization is where we’re aiming for CHLOE to expand. CHLOE is CHLOE EQ, which stands for Embryo Quality. There are different CHLOEs. Some of them are already in proof of concept. Some will be developed in the future that will expand CHLOE’s reach into more areas of the IVF journey.

Now, we know. We will be watching carefully to see how CHLOE is doing. This is incredibly exciting. I’m glad doctors aren’t going to be replaced, but rather will be helped out with this type of AI. Thank you so much for joining.

 

BOV 69 | AI For Fertility

 

It’ll make all of us better.

It was a pleasure to speak with you.

Same here. Thank you so much, Alyssa and Rachel.

Thank you so much, especially for taking the time. We will be watching and cheering you on because this sounds like it will help patients, physicians, and ultimately families.

Thank you for having me.

 

Important Links

 

About Erin Eshed

BOV 69 | AI For FertilityEran Eshed is the CEO & Co-Founder of Fairtility, the transparent AI innovator powering in vitro fertilization (IVF) for improved outcomes. He is a multidisciplinary business executive and serial entrepreneur with over 25 years of experience spanning numerous product and business domains. Eran was a Co-Founder and Chief Business Officer of Altair Semiconductor, a wireless chipset innovator acquired by Sony in 2016. Eran holds a BsCEE in Electronics Engineering from Tel Aviv University.

Leave a Reply

Your email address will not be published. Required fields are marked *