Latest / AI Meets MedTech / Reinventing Maternal Care with AI & LiDAR
Transcript
- speaker-0: This is AI Meets MedTech Podcast with Anna Nestorova. Welcome to AI MidsMetTech. Today my guest is Kamola Meer, the founder and the CEO of LightVx, a company with a big mission to enhance maternal care. Kamola, welcome, and I'm so glad you're here. speaker-1: Hi Anna, thank you so much. I'm so pleased to be here with you too. speaker-0: Well we're gonna start right away. I wanna ask you a very important question. What is one thing that women experience during labor that most of people have no idea is broken? speaker-1: ⁓ the one thing I would attest to and which is what we are on the mission to fix is ⁓ the labor monitoring process, how we observe the labor progress, how we basically assess ⁓ the types of medical interventions we need to make based on how the labor is going on. Unfortunately right now it's manual and it's done by different types of providers at dur ⁓ during different shifts. So it might be one person doing one type of manual exam, the next person comes in with the next shift doing it differently or obsessing ⁓ observing it differently and unfortunately usually that gives ⁓ inter opera ⁓ interreliability issues usually before between the providers leading to ⁓ incorrect assessments, incorrect ⁓ data, and obviously incorrect labor management that was maybe not even needed, different kind of interventions, etcetera. Unfortunately that is broken in the current medical system all around the world. speaker-0: Yes, I was gonna say, and you think we've been doing it for thousands and thousands of years and yes the manual assessment is the only one we can rely on and nobody has done much to f to change that. So tell me how are you trying to fix that and the technology you're developing, how is it changing the way it's done now? speaker-1: Yeah, first of all I would like to start about the anciency of the problem itself. You would think ⁓ since it's ancient and it's been done for many millennia, why fix something that is not nearly broken to the point that it stopped the humanity from reproducing? ⁓ and that was our mission to not fix something, but to make it more fit for the modern world to the modern era of data, to the modern world of care providing. that is more accessible to different layers of wap population. Because if you look at different layers in ⁓ underserved communities, in different locations where they might not have enough opiology experts to talk to, to have some care from, those people for millennia didn't have enough opportunities to have safe labor. And it was always a coin flip whether a mom dies or no. We want to bring the care to these places instead of saying to the doctors, hey, we are fixing what you don't know how to do. In that terms we decided to bring something very simple to use, a handheld device called Servilite, which can potentially now right now it's through going through the clinical trials, can potentially ⁓ provide the clinical data net necessary during the labor by using LiDAR technology. It's technology that is used in VAMO, self driving cars, in our iPhone cameras, etcetera. speaker-0: So let's focus on that. So LiDAR is a technology that allows self-driving cars to see the road, correct? Mm-hmm. So how did you come up with an idea to point it on cervix and then make a machine read the basically understand the reading of it? That's fascinating and I wanna s point it out because the whole podcast is about technology. So tell me more. ⁓ speaker-1: No. Thank you for asking that point. Actually, that was ⁓ one of the main points why I came up with the technology. I was looking for different kind of solutions for cervical observation, for diagnostics. And when I was looking through the current medical technologies, I didn't find something that could be used for that specific purpose. Then I went to Home Depot to look for construction stuff that use that is used for ⁓ scanning the sites. ⁓ and obviously when you look through different kinds of technologies, you find something that might not be used in medical field and it can be utilized now. ⁓ unfortunately I didn't find anything that was irrelevant to me. Then when I went back home ⁓ and I was talking to my husband about ⁓ birthing experience, etc., all of a sudden my Roomba started hitting the wall and I was like, I think its lidar is broken because it it's probably not scanning the room correctly. And that hit me. I was like, ooh, Eureka moment. It's a moment of basically having this light bulb in your brain. LIDAR, why not you to use it in the body? I started searching the patents, checking how it's usually done. I ch started checking the safety regulations in terms of is the optical technology even viable in the human body. And all in all, different speculations, different specific specificities ⁓ led me to choosing the lidar and finding the lidar that is actually appropriate for the human body and patenting it for the utilization for anatomical measurements. speaker-0: That is a fantastic story. I want just to maybe this there will be a great short clip for our interview, how you go from the a Roomba Robot to the medical device that it can be used in in the body. That's a fantastic story. So tell me a little bit more about the technology behind it. So you are you using AI with your device? Is it part of the ⁓ development? speaker-1: ⁓ So we are a company of the maternal health infrastructure. So we have a medical device that is purely a medical device. Basically, it has a software component that uses SliDAR for algorithm calculations. But it we also have a separate ⁓ product. One is B2C app for mothers to help their ⁓ calculation to when to leave to the hospital and with labor navigation basically throughout the labor process and beyond. And we also have ⁓ the upcoming product ⁓ that utilizes servic server light data. To give predictions with AI for labor outcome management, for additional scans for cervical oncology, fertility, menopause. We think that data is the most valuable component right now in the world. And AI, any AI in the world is as good as the data it's trained on. Given that Femtech doesn't have enough data, especially in the cervical area, we think that this would be the most valuable asset for as well. speaker-0: Yeah. And when it comes to AI, I want to ask you so what was not possible before era of AI with something you're building right now? And what is possible now with availability of AI? I'm just trying to uncover what has been changing lately. And obviously everyone is talking about AI so much. We know everyone is talking about AI in different fields, but specifically in your developments. So what was not possible before AI was implemented? speaker-1: For LIDAR itself, I wouldn't say AI was an essential component for the medical device itself because ⁓ LIDAR is physics, we are just using optics. Now for the B2C component and for the data component, we cannot ⁓ do anything with the data we obtain without AI. All these ultrasounds, everything, they are used for the point diagnostics, right? To diagnose a woman right now here, to give the predictions here or anything. But you cannot use it for population is population study. You cannot use it to study all these patterns among different thousands or millions of people. And AI is the only tool that enables you to be able to observe millions of women at once, get the data and get the patterns very quickly. Because before that we would spend tens of s ⁓ years, I think decades to be able to analyze one million patient data. speaker-0: Would you say that's the main misconception that people have about AI? Because I do find that sometimes people think AI is something like robotics, like there's a robot who thinks instead of a doctor or a nurse. But do you think that's the biggest misconception? The data processing, that's what AI allows us to do in medical. speaker-1: I do think so because right now everybody, especially in the general population, not in the professional population, general p publicity thinks that AI is mostly about chatbots, about talking to someone and who it's better better way of Googling basically. And that's not true. AI is so massively underutilized in other areas. Palantir, for example, the biggest companies who ⁓ analyzes di data for the governments, for ⁓ private companies, etcetera. they use AI for data analysis because there is so much data all the time coming in, coming in, and you don't know how what to do with that data. And there is so many opportunities, there are so many ⁓ ways of using that data for for the humanity, for better health outcomes, for better ⁓ governmental efficiency, for anything, etc. So AI is definitely a massively underappreciated in the data area by general publicity, and I think that's the c most common misconception as you mentioned. It speaker-0: What was the most surprising thing about AI that you discovered during your work with the device, with the app that you're developing? Is it something that surprised you the most? speaker-1: How accessible it became. I didn't know how accessible it is now to people even without engineering degrees. For example, I don't come from engineering or computer science degree, and my only coding skill is basically biostatistics, anything relevant to research and biology. So I thought I would not be able to comprehend the level of skills that you would need for AI ⁓ AI work, basically, for creating your own neat AI native company or anything. But now AI is so much more accessible. You can do it with any ⁓ skill, without any technical skills. You can still ⁓ learn AI, know how to use it and know how to even create it. speaker-0: This is AI Meets MedTech Podcast with Anna Nestorova. Mm-hmm. And go back to your technology and both device and app. So do you think it may have some pushback from healthcare professionals in a sense of, ⁓ it's a a new method, completely different. Like you said, we've been doing what we've been doing for thousands of years. Why change it? Like what is your expectation? Do you think you may have that? Maybe you've already experienced. speaker-1: Yes, that's a great question because we did have a pushback ⁓ initially. That's why the company didn't start immediately when we patented the device. The patent was ⁓ happened even before the company incorporated. ⁓ and the first thing I did was pretending to be a medical school st student and going to Shadow at Providence Hospital. And what I did was observe the workflow of the ⁓ local Obigane colleges because by my experience I'm an Obigane College nurse from Uzbekistan. And the workflow there might not have matched what they do here. And I saw that most of the stuff is actually very similar. So there was not much workflow difference. But once I started questioning the vaginal exams, the manual exams that they do, there was a pushback on ⁓ having a device because no doctor wants to admit that they are doing wrong. It they always say, Yeah, there is a statistic that forty-nine percent of the exams are inaccurate. But It's not me. It's those rookie students, student residents who are doing it incorrectly. I don't have that statistic. ⁓ and that's every single obignecologist that says you says you about it. But then you see the studies from Harvard publication where all these experienced medical ⁓ providers are doing all these exams and are doing it incorrectly. So what we did is we didn't want to fight them. We don't want to fight the decision makers because they are the ones ultimately who are buying it and the hospital Workflow system is basically decision making doesn't only come from the doctors, it only also comes from the finance office, right? And the finance office, maybe we can convince them that there is a huge ROI. But then that if the doctor doesn't want to adopt it, no matter how many devices they buy it, it's not going to be integrated in the workflow regardless. ⁓ so what we decided to do is pivot into the underserved communities where the need is like the most essential, the most obvious, and where it's mo the most agreeable, so that the doctors don't have a pushback on that. Underserved communities are right now one-third of the counties of the United States. They're called maternity deserts, places they don't where they don't have enough obiganecology experts, ⁓ meaning that over 2.3 million women of the fertility age, they don't have enough access for safe childbirth experience. And that actually, the being in the maternal deserts, ⁓ increases the risk of maternity death by 36%. It's a huge statistic right now. The transfer rates are 2x because the providers there Who are not trained in obignecology, don't know how to ⁓ observe the labor, how to manage it, and don't have the capacity to make the decision on time. And recently the Senate ⁓ issued a bill on maternal health called Momnibus, ⁓ Black Maternal Health ⁓ Omnibus. That one was basically issued about increasing, improving the telehealth communications for rural areas, where, for example, a provider, not necessarily obey gynecologist, but any kind of provider, can do some kind of technological assessment on a woman and telehealth communication sends the information to the to her main doctor to the obiganecologist in the bigger hospital where she can make the decision instead of the provider with no experience has to make the decision on time here. So that's where our device comes in. We are the extension of the hands of an obignecologist in the rural areas basically. Yeah. speaker-0: That's fantastic. I think it's such an important question to be, I think, discussing overall. And it surprises me that not many people talking about it. So that's fantastic what you do. I have a question about that extension. So there's a doctor who's sitting is sitting in the main hospital. There is someone who's maybe less trained on this. But do you think with the power of technology and potential AI, you can train those people quite effectively so they can use the device properly. that translates information to the doctor because I would imagine that that clinician would have a very important job to do, to make a decision remotely. Even though it's happening right now and I talk about it quite a bit with remote surgeries, et cetera. But it's still a very it's a big shift yes from how we know the system is working and from how it it's gonna be working in the future. So do you think with the power of technology in AI you will be able to train people quite effectively remotely like that. What's your take on that? speaker-1: Yes. ⁓ so we interviewed a few family physicians who are already doing the medical assessments in the rural areas. So they still have to do the manual exams. They're just not experienced enough to do it properly. So these kind of ⁓ providers, they're not uncomfortable with using a technology that is new because they are not absolutely changing their workflow or adding additional exam. They're replacing the current exam with something that is extension of their hand, basically, instead of having to do the manual assessment and not being sure about it. They now have an assurance through the device. And that actually also goes into the medical liability. ⁓ they don't have a liability on them. The liability is put on the h on the device because the with the device you can prove why different kinds of mis d decisions or transfer decisions were made, medical interventions, etc etcetera. And you can always go back in the history of the device memory to see what kind of exams were done, ⁓ what were the measurements, versus with hands you can never go back in time and check. how it actually measured versus what was written on the paper. speaker-0: Yeah. So now I'm looking at it from the patient's perspective. And as a mom, obviously we all you and I, we experience that. So now I can potentially obviously we're not talking about your product as you said, because it's not available ⁓ in a commercial sense, but I w should be able to educate myself better. So when I go and see a doctor, I should be able to ask those questions. And actually that's the purpose of what I'm trying to do here too. So people can be more educated and maybe they can ask right questions. How I'm gonna be assessed. Because we do think, like you said, we do think it's Normal, we don't know sometimes, but we now raising awareness, I guess, how to manage the labor better. And with your app I believe you also you put it in the hands of mothers too, so they can actually observe and see what's happening with their body. Is that correct? Am I understanding it correctly? Yes. speaker-1: Totally. It it's actually the reason why I started this company because ⁓ during my first childbirth experience I didn't have enough control over my own childbirth experience. I didn't have enough data in my hand to observe what was going wrong when things went wrong. And thankfully ⁓ everything ended up well, but there was some tragic moment where we were almost about to lose the baby. So after that I decided that next time I don't want to have this same experience. I want at least some ⁓ amount of control. over things that are happening around my body, around my childbirth experience. And that is why I started this company to bring control back to mother's hand. So that we at least are informed, even if not capable enough to make decisions, because usually the medical professionals are pretty convincing at the decisions that they're making for your own sake. But we at least know what is going on and are informed about the context behind it. Yeah. speaker-0: Yeah, and I think that's an important part too, so because we're not trying to push all the responsibility to to a patient's hands, obviously. But it's nice to be much more aware. Yes. Versus being blind when you go blindly. So I do agree with that. Because I do appreciate the work that healthcare professionals also do. And that's part of where I'm asking about the pushbacks and how healthcare professionals are also can be quite conservative in the way how they change their practices, which I can also understand. But I do agree that we need to have a little bit more education in Capability, especially when it comes to labor, when you do spend some time on your own before you get to the hospital. So that you can start tracking things before you go exactly, right? Awesome. This is AI Meets Tech Podcast with Anna Nestorova. I do have c another question about AI. So is there is there anything that surprised you in general in healthcare, AI and healthcare in the recent months? And we're not even talking years anymore. We're talking about months and weeks now. Is there anything that you can share with me that surprised you, you found fascinating? I just like to hear stories maybe that I I haven't heard. Mm-hmm. speaker-1: ⁓ I was just actually watching one of these influencers online. She is the spouse of one of the founders of Hicksfield AI, the Kazakh ⁓ Unicorn AI company. And she was talking to one of these founders in AI who developed wearable that scans your brain to understand your focus, your different workflow patterns, to understand where when you are the most efficient, productive versus when you need to sleep, when you need to focus, etc. And obviously you can say that. Apple Watch can also do that, or Aura can do that. But at the same time, observing your neural patterns is even more effective. Especially for people who only work with brain work instead of working physically. Sometimes maybe their ⁓ bodily response is not changing, the heart rate or temperature is not changing because they are sitting still all the time. But then the stress of the or amount of work might be creating these neural patterns that are not visible to normal eye or to normal verbals. So I think it's fascinating what they're doing now with the neural work and with all the productivity tools for the rain. speaker-0: That sounds really interesting. And is there anything that you would find maybe a little bit concerning about AI and healthcare? Is there any anything on the it can be on both professional, personal level? Because I know some people experience that some rate of anxiety with AI being around them all the time now. Yeah. So is there anything that you can think of? speaker-1: ⁓ I would say probably again going back to data, how much of data is actually supposed to be at the hands of AI and what is what is it used for? For example, there is HIPAA issues, there is protection protection, there is some privacy. And I would probably assume that having too much data for the AI, for the sake of AI is also a little concerning. ⁓ at the same time giving bac giving too much ⁓ control to the AI in the workflow, for example, I don't know, integrating A AI in the whole EHR integration instead of having portion of it would probably be concerning because now it has access to everything and can accidentally wipe out everything. Like for example, the ⁓ what was that Claude's recent version of the AI that they ended up closing? ⁓ that one was having some concerns. These kind of security issues are a little bit of concern, especially because we don't have any security protocols for AI yet. We only have protocols for the general coding. Yeah. speaker-0: I do agree with that. ⁓ if we're talking about labor, going back to the main subject here. So if ⁓ a woman is in labor, either at home or going to the hospital, in the future, in the ideal world, with the technology that you're developing and overall the technologies that we have available now and developing right now, what would be your ideal scenario? How do you envision future experience of a woman going through labor? You can go speaker-1: You can go I know, right? I'm already having some visualization in my brain, to be honest. speaker-0: And just dream it's fine too. Albody's done so much actually, you're on the path actually of changing how it can be. So so what's the ideal scenario? speaker-1: Ideal scenario would be probably first of all, brought around the comfort of the woman. So focus on obviously safety, which is right now the main protocol, but comfort being at the same level as safety, so that women are ⁓ feeling secure about their childbirth experience, are informed again, as we mentioned earlier, but are also comfortable enough to trust the medical personnel. I don't have to feel like they need to be in control of the situation because right now the reason why we think that we are underinformed and have to be in control and have to be sane all the time is because there is trust issues from time to time because of malpractice in the medical field. Obviously it's not about all the doctors or anything, it's about the particular situations where women get hazardous and obviously end up being more alert in the burden situation. But the ideal version would be where women are comfortable enough to trust the process. ⁓ the whole room being equipped enough but not too much ⁓ so that there is not much anxiety adding up for additional monitoring. I would rather have one monitor that can do a lot of things that multi than multiple monitors going down through the transvaginally and one going on your belly, one going on your wrist. That's too much. We already have enough anxiety during labor. We don't want to have so much restriction, especially when you need to move around during contractions. You need to jump on the yoga bowl, you need to soothe yourself, you need to sing, hum to yourself, right? Breathing techniques, everything. But you get so much anxiety from having so much stuff that's attached to your body. ⁓ and I think ideal version would be just having one monitor that can get all the data needed and then everything being visible to you in a normal format and then in a clinical format to the provider that is actually observing your birth. speaker-0: Mm. No, that sounds great. And maybe the last question is what is your wish personal for AI? If you would have a magic wand, and now we have AI, so what that magic wand would do for you to make your life better, easier, I don't know, anything that could help you to make your life in some sense better. Yeah. speaker-1: based on the fact that right now we all have AI agents, I would rather have AI personnel rather than AI agent because AI agents still require so much work to do around them to be able to be utilized correctly. And from what I'm learning from all these companies that are utilizing AI, it's costing more than what it's bringing. ROI is lower than the actual loss that they are having because of switching to AI for the sake of switching to AI. So my ideal AI would be something that is so automated that I don't have to do anything or worry about anything. Maybe learn from my phone, my patterns, my behavior, anything. Do whatever you want to do. Here's my phone. Here's my phone. Spy on me first. We are already spied, right? We already don't have any any privacy, regardless. Mera is I wouldn't say spying. It's collecting data. Everybody's collecting data. Just use that, please. Just train your personnel so that it's a perfect assistant instead of me having to train it and constantly remind it to be more honest instead of being this people pleasing chipity kind of AI. speaker-0: I'm the same, I actually tell my squad. Always please stop doing that. Thank you so much for having me here. It was AI Meets MetTech. Follow Kamola. We will share some information about her company, about your LinkedIn profile. Are you okay with sharing that? Of course. LinkedIn profile. And if you have any questions, please feel free to shoot her a message. Thanks again. speaker-1: Brutal honest brutal honesty. ⁓ Thank you so much, Emma. speaker-0: This is AI Meets MedTech Podcast with Anna Nestorova.