Sonia Grego (Co-Founder at Coprata): Oura Ring for sleep. WHOOP for activity. Now, Coprata Smart Toilet for stool
Sonia Grego (Co-Founder at Coprata):
Oura Ring for sleep. WHOOP for activity. Now, Coprata Smart Toilet for stool
It’s something that we take for granted — until we can’t. How patients communicate about stool is anything but objective. What’s missed in translation is the ability to diagnose and treat GI conditions in a timely manner.
Therefore, it was with much intrigue that I began my conversation with Sonia Grego, PhD. She’s a professor of electrical and computer engineering from Duke University, the founding director of the Duke’s Smart Toilet Initiative. Her team worked for eight long years to develop a smart toilet. They spun off the innovation as the startup Coprata that has won competitions at Harvard. The Coprata toilet captures lots of data post flush, runs AI algorithms on the data and provides the analysis to both patients and doctors.
One, the market is showing us repeatedly that many new innovations in GI are coming from outside of the specialty. As a gastroenterologist, you must consider that as a missed opportunity.
To stay relevant, it’s so important to stay up-to-speed with what happens beyond the endoscopy room.
Two, increasingly new devices are preparing the ground for digital-first GI care. The smart toilet can go in many different directions from here. With scale, it can create so much ongoing data that it has the potential to change the trajectory of how much GI conditions are managed.
Praveen Suthrum: Sonia Grego, welcome to The Scope Forward Show. I’m very excited about this interesting topic. We’re going to be talking about smart toilets. So, I first want to welcome you, Sonia. Thanks for joining us today.
Sonia Grego: Glad to be here.
Praveen Suthrum: So, Sonia, I want to get started by sharing a little bit about your background. Sonia Grego, PhD, is founder and CSO of Coprata and professor of Engineering at Duke University. She has led the development of a smart sampling toilet that automatically captures stool data post flush, and she’s passionate about bringing this product to GI patients to improve outcomes. Very interesting background, I must say. But I want to start by asking you, I’m sure growing up or while studying engineering, you didn’t dream of working on smart toilets. So how did this come about?
Sonia Grego: It’s been an interesting journey. So, my PhD is in physics, and I have over 20 years of experience working in applied technology and engineering, developing biomedical technologies. I worked on wearable sensors and biosensors in a variety of capacities. I started to be interested in toilets a few years ago, I would say eight years ago. Myself and colleagues here at Duke University, we have an entire center that has had large programs funded by the Gates Foundation to develop toilet technology. If you think about it, toilet is a fantastic appliance, very effective in doing its job of removing waste. It has not changed since it was first introduced in the home in the beginning of the century. It’s a white ceramic bowl with water and you flush it, and your waste goes away. And it is fantastic for that. As we were working on technologies for treatment of the waste, we posed the question, and particularly my interest in biomedical technologies to have an impact on health. We wonder, well, is there data in this waste that we are flushing away? Can we capture it before we flush it away? And the answer is a resounding yes. And that is what we set up to do, technology that analyzes school data. Also, I want to add in my experience with sanitation technologies for other environmental applications. We have deployed and tested many toilets with real users, and we become aware of how sensitive the topic of using the toilet. It’s a private, personal act and users, and particularly women are very sensitive about the use of a toilet. So, we had designed a product for the Coprata smart toilet. We went really great engineering effort to design a product that does not appear different to the user so that it doesn’t engender discomfort.
Praveen Suthrum: I just want to delve a little bit deeper. Was there a certain trigger? How did the exact idea come about or were you just simply thinking, hey, we got to measure stool data? There is a lot of data and stool, so let’s figure out a device or make a device that helps us do that. What was the process there?
Sonia Grego: Well, the process was, we know and speaking with physicians, collaborators and gastroenterologists here at Duke, we learned, yes, there is a lot of data in stool, and it is very difficult to get it. Gastroenterologist collaborators told us that they spend most of the visits with the patients just figuring out what was the regularity in the bowel movement that is associated with the concern that brings them to the visit in the first place. They say, well, 90% of the visit is just figuring out the consistency. What do they mean, the volume? Is it little? Is it a lot? What little? What’s a lot? People do not have a frame of reference because it’s a private fact and everybody only knows what they are doing. The physician told us, yes, I would love just to know how much they tell me. They go to the bathroom 20 times a day. Is that really true? There is something coming out really 20 times that seems impossible. And yet there is this mismatch between what patients report and what clinicians understand. So, there was data in that. And then additionally, they also said, well, and when you need a stool sample, you send them home with a stool kit and you get it back. Hardly. Maybe 40% of the time, people just don’t want to do that. So, the need was explained very clearly to us. And we said, well, of course, you could just engineer something that scoops down in the bowl and picks it up or takes a picture of it from the bowl and you’re done. But that’s where our engineering experience said no, I mean, you can do it, but if you can do it, you should not do it, because nobody will ever use it. Our experience is everything has to happen in a system that appear normal and usual to the user. Toilet users, they just want to see a white ball, no gizmos around it, water in it. That’s it. That’s the only thing they want to see. So, he said, well, in our laboratory we are completely set up to test toilets. We have them on laboratory benches. We are very familiar with the physics, the fluid dynamics, the engineering of the whole system. We said, okay, let’s figure out how we do this stool analysis after the stool has left the bowl outside the purview of the user. Well, that’s easier said than done. That was months of brainstorming and test and tried and truth. So, our current technology is the result of a large number of tests and failures. But now we have it and we think our approach is unique and it is possible.
Praveen Suthrum: Can you explain to me what exactly it does and how it does what it does?
Sonia Grego: The principle is all toilets are designed to take the waste and move it to the switch line as fast as possible. So, you flush, and the hydrodynamics is designed, fluid dynamic is designed in such a way and the stool moves very quickly down the drain. The point of our invention is we were able to immobilize it for a brief moment and in a reversible way. So, whatever happens in the pipe to the stool after it leaves the ball is it stops in a region of the plumbing. For a brief moment, all the sensors are placed, and we get the information that we need and then the stool proceeds being flushed away. So, we have a toilet that is designed to do sort of the opposite of what regular toilets do. Yet the appearance of it to the user and to the customer is of a very regular toilet. We could show you a picture from a toilet from Home Depot, you could not tell the difference. Which is the point. And we think that will really facilitate adoption because people, for example, this is designed for residential use to be installed where people go to the bathroom, which is at home, in case they have family members, they don’t want to be tracked or a guest coming in the home. We want the toilet to look like a regular toilet, but yet being able to capture all this data.
Praveen Suthrum: So, this is the entire toilet itself. If a user has to use it, they go to replace their existing toilets.
Sonia Grego: Yes, we understand that’s something that people do not often change their toilets. So, this is a great opportunity. The installation is provided as part of the product is a 90 minutes operation and the toilet that is removed, we have a reuse of it. We crash material for construction, so we have a sustainable approach to the reuse of the old. We called them unintelligent toilet. The user has to install a whole new toilet but will make the experience as seamless as possible and that is a one-time operation. After that they will have a toilet that operates like a regular toilet, but as an opt in solution. The member of the family that wants to be tracked, they will use the toilet as they regularly do, but then on their phone app will appear data trends that describe the values and the information that our senses and algorithms produce.
Praveen Suthrum: I want to talk about the sensors. What kind of sensors are there and I’m assuming you’re taking a picture also. I’m curious, how is that possible? Now, is the toilet plugged in to an electrical socket or is it battery operated?
Sonia Grego: The system requires power the same as conventional smart toilets that are on the market. There are products that have features such as heated seat, incorporated bidets that require power. So, this system does require power. So, it will be either a socket if that’s available, or power operated, or battery operated with a rechargeable battery like your power drill. And in the absence of power, it does not record data. That’s the only thing that can go wrong. It still works like a regular toilet unless a signal is given, and the data is collected. So, it just operates normally. Indeed, there is a camera, and let me specify again, the camera is in the plumbing. The background image of the camera is a piece of pipe. So, there is no concern whatsoever that the user or the bowl is ever involved in this imaging. So, we capture images of the stool, and we have extremely high-quality images. We also have other sensors that are commercial off the shelf sensors or sort of customized physical sensor devices. So, we are able to capture parameters of the store properties, which are what clinicians typically ask of patients. So, whether it’s a urination or a bowel movement, what’s the consistency? Both from images as well. We also have a specific diarrhea sensor that measures the turbidity of the wastewater to capture the component of a liquid stool and that allows altogether to have a complete. We have the full range of Bristol scale from one to seven unusual colors and presence of blood. And importantly, also sitting time and the duration between your first sitting on the toilet and when excretion occurs. Because that we believe is a surrogate of urgency or straining. Which are parameters that clinicians are interested in knowing about and currently are just qualitatively expressed.
Praveen Suthrum: How does this data, I’m assuming this data gets transmitted either via Bluetooth or the cloud and go somewhere. So, tell us that process.
Sonia Grego: So, imaging, and analog sensors, and these are operational, and our first pilot is ongoing in our own facilities. So, we are collecting the data as we speak. The data is collected and analyzed on our own servers, and it is coupled with algorithms so that all these signals become actually information parameters. So, people will get a score about their sitting time and their changes from a baseline.
Praveen Suthrum: So, Bluetooth. Or is it Wi-Fi enabled? How is it getting?
Sonia Grego: We can configure this. Made it at the moment is Wi-Fi enabled just for convenience.
Praveen Suthrum: Just like a smart TV. Now there’s a smart toilet like you configure to your home network. All this gets analyzed and relayed to an app.
Sonia Grego: Yes, we envision the data as summarized enough in form of a dashboard for the convenience of the user. So, one of the questions that we always receive is once the toilet is installed, these sensors will work indefinitely. So, it will capture pictures of every bowel movement. One user, one bowel movement per day. If we had 100 toilet installs, we would have 200,000 data points in a year. So we believe our Coprata Toilet will truly be an at home bio sensors for GI tracking that so far has not been developed and the data clearly needs to be summarized in dashboard, both for the users and for the clinicians that are interested in seeing it. So for the users we envisioned an app, and for clinicians, some form of easily transferable information.
Praveen Suthrum: I’m assuming that the toilet is designed for a single user. What if more than one person uses the toilet? Does it differentiate between one person’s stool and the other?
Sonia Grego: Yes. There are many ways in which the toilet can differentiate the user and we envision that customizing on the customers. Younger clientele are happy to just have the toilet recognizing the cell phone that is approaching the toilet. For older users, people have asked us can I just have my own button? So when I use my own flush button, I’ll flush that button. So there are many ways in which the toilet recognizes users, but I always specify that will be customized to the client and how many people want to be tracked. And people that don’t want to be tracked will not be. So It’s an opt in system.
Praveen Suthrum: While the development of the product is going on. How does one even test for something like this in a lab or at the university?
Sonia Grego: We certainly tested extensively on laboratory benches before we ever installed in a bathroom. And we have here at Duke University a unique facility that was designed to test toilet. So we collect specimen donations, stools from healthy volunteers and pour flush in the toilet. And that’s how the science is being done for development of the technology. So it’s a unique capability, but we happen to have it and we are very confident of the quality of the product we have developed. It comes from a sort of deep bench of knowledge and capabilities in this specific space.
Praveen Suthrum: Do you have a number in terms of how many data points that you’ve collected so far? What kind of data have you collected so far in terms of, I want to talk about quantity. I understand the quality of data. I just want to curious about the quantity of data that you have.
Sonia Grego: I don’t have it off the top of my head. We have developed an algorithm on 3000 images. Those were obtained by crowdsourcing. We have published a study on the ability to sample the stool ones from the toilet. And there we had hundreds of data points and for the toilet that we had installed. Now we have just a few months of data, you can calculate multiple up to five users per day, times seven times a few months.
Praveen Suthrum: What have you learned so far?
Sonia Grego: We have learned from a stool image. An algorithm is very good at recognizing consistency and as reliable as clinician. There are studies in the literature, since this is an area that is being investigated, other studies have reported that algorithms are better than people at recognizing the specimen. We are really comfortable with the assessment that sensor-based analysis of stool will provide clinicians more accurate data than what a patient could do, just turning around and looking. We have also learned that it is possible to sample the stool from the toilet region and it’s possible to conduct biochemical analysis on that sample. This capability is not in our product yet. It has been characterized separately and with funding from the NIH. So our platform has like near term ability to collect all the information on the bowel movement, but it will also be developed longer term towards stool sampling and having a fecal specimen that can be sent the Laboratory for Biochemical Analysis.
Praveen Suthrum: This is the earlier version of the product, but once you get to later versions of the product, what do you hope to achieve in terms of the science aspect of it? From the scientific aspect?
Sonia Grego: We think the Coprata toilet will be first tool that empowers people with their data so that they can take better care of themselves, to have longitudinal data about their bowel habits, so they understand when a lifestyle changes, makes any fact. They will empower clinicians to reduce uncertainty in prescription and management. In my conversations with Gastroenterologists, they always say, well, I’m trying this, I’m trying that. They tell me it’s not working out. I’m surprised that the patient gives me this feedback that I feel like it’s almost impossible. I feel like there is a great uncertainty. I hear clinicians talking in follow on conversations with their patients, looking for potential alarm signs that the diagnosis was not right. So, I feel like that especially in many chronic GI diseases like IBS or disorder brain interactions, there is so much uncertainty from the clinician and the patients. Our dream is to put this uncertainty to rest, or partially at least, with a steady stream of information that rules out the alarm bells. I was told, for example, that bowel movement, nocturnal bowel movement, are an alarm time. I’m like, wow, so something just like a timestamp when you have a bowel movement could tell your physician that, wait, there is something wrong, the condition is evolving, and we need to do a different treatment. That’s something that our tool would be able to do. And give peace of mind to caregivers for caregivers of children or vulnerable populations that cannot speak for themselves of what’s going on in the bathroom. So we want to give patients and clinicians certainty and value from the data much longer term. We envision that the stool sampling capability will be added to our platform and that will have to be an FDA clear device. So, the development timeline is a bit longer for that one, but that will empower, for example, celiac patients to check for inadvertent consumption of gluten in their stool, to conduct microbiome tests, to conduct stool tests for patients with IBD, or for patients for which fickle test is recommended multiple times a year and is not a pleasant act.
Praveen Suthrum: What kind of disease conditions are you currently planning for?
Sonia Grego: The toilet would benefit patients with all sorts of GI conditions that result in power regularity, which is practically all of them, but particularly the IPS functional constipation people type of an IBD patients would immediately benefit from the toilet. We also think that health-conscious people, even people that are not currently patients of GI but, any health-conscious person that has an Oura ring to track their sleep, that has a whoop to track their physical activity, would benefit from tracking their bowel movements so that they see what their diet and lifestyle does. I mean, regularity and gut health is health and happiness for the person. We also envision used in of course research studies and clinical trials. Like clinical trials, particularly for GI conditions, they rely on facial self-report for many of these improvements in bowel movement frequencies or straining or urgencies. And we believe our technology would provide investigators and stakeholders with a robust set of data for that. So, the sky is the limit. There is a lot of white space in the toilet monitoring your gut health. The reason being we believe is very difficult. What we have developed is technically very difficult to do because of the heterogeneity of the human stool. While urine, for example, early on in the development of the toilet, people asked as well are you doing a urine analysis toilet or are you doing a stool analysis toilet? And we reflected and our choice was very conscious. Urine is the liquid is not technically that difficult to have a urine analysis toilet and capability could be added on to our platform. We know exactly how we would do that. Stool analysis toilet is much more difficult because anything can come out of there from very hard to completely liquid and you have to capture it all because particularly the extremes are of interest. So, we tackled very deliberately, head on a very difficult problem. From an engineering standpoint, we believe we have solved it and now we are interested in having a product that will bring benefit to patients very soon.
Praveen Suthrum: Let’s switch gears to the business aspect of Coprata. So, what can you share with us about where you are as a startup, as a business? Where is the company currently?
Sonia Grego: The company is spin off of Duke University. We started with pre seed investment by enthusiasts and believers of smart toilet that have been supported our research and they nurtured it since it was at the university stage. We are making fast progress. Our team is growing headed by a CEO. We have completed pre seed round. We have raised funds from federal sources and other sources. We have won first prize as a new business venture out of Harvard Business School. So, we have received recognition for the originality and the potential impact of our approach. And of course, we are now raising funds again.
Praveen Suthrum: What is the product price at do you know already or is it still evolving?
Sonia Grego: We are still working on the pricing structure of the smart toilet, but it will be in the range of the existing smart toilet.
Praveen Suthrum: What is the business model? Is it mainly by selling directly to consumers or are there other types of revenue sources that you’re envisioned?
Sonia Grego: Our first step will be a direct to consumer. We believe that once the users expand and we have a base of users, we’ll be able to conduct studies and demonstrate the value to healthcare of the data that the system produces. So, once we are able to demonstrate the improvement in outcomes, the savings in time and healthcare costs that the data produces, we envision more of a B to B to C model in which payers and big employers will be interested in subsidizing or partially reimbursing the device for their patients the same way as it is done for diabetes. Now, diabetes is a chronic disease is far ahead than GI chronic diseases, but GI chronic diseases are pretty expensive and really impactful on the quality of life of a large and increasing number of people. So, we believe that large employers and payers will take notice of a system that enables remote patient monitoring with all the savings that come and then further out, our market can expand to stool sampling and medical grade devices that target specific population and the whole sampling analysis. So, our model starts with a clear focus, but it expands into many directions.
Praveen Suthrum: How far are you from a public launch of the product?
Sonia Grego: We are recruiting users for a pilot launch of our product later this fall and we want to have products for sale from our website in 2023 next year. So, we are very close to the launch of our first product.
Praveen Suthrum: One final question before I let you go. Given this journey, in your view, what is the future of a specialty like gastroenterology?
Sonia Grego: The future of gastroenterology, like of many other specialties, is in telemedicine. That’s what called it as push forward. And digital technology has developed. For this specialty, particularly where there has been no at home biosensor tracking specific physiological activity of the patient. We believe our toilet will be an important tool that will enable care that is remote, that is proactive and not reactive, and that will track patients in the long term to keep them in a state of remittance and health.
Praveen Suthrum: It’s quite an amazing innovation. I’ve never spent so much time talking about stool, leave alone a toilet. So, it’s been very insightful and it’s amazing to note your journey from biomedical engineer to the founder of Coprata, Sonia Grego. Thank you so much for joining today on The Scope Forward Show.
Sonia Grego: Thanks for having me.
Praveen Suthrum: It was great talking to you.
COVID-19: The Way Forward for Gastroenterology Practices
COVID-19 is a double whammy of both clinical and business disruption. This ebook will help you explore possible scenarios and be a guide in your plans for the future.