Inside MEI Micro's mission to bring military-grade sensors to healthcare

0:03So we have a slogan that we use a lot. You can't software your way to a better sensor. If I use a navigation grade sensor and I build my software based on this high accuracy data, the end result is going to be objectively more accurate.

0:22Louis Ross, founder of MEI Micro. I appreciate you joining us here on Startup Health. Thanks for having me. You're working on this sensor technology, typically known for use in accelerometers and in navigation. But obviously, we're here with Startup Health talking about the application of these sorts of sensors in healthcare. So excited to hear about what you've built and why you've built it. So why don't you just start by giving us an overview of what makes MEI Micro unique? We've been working on our main process technology we call 3DS. It is used to develop the world's first navigation-grade multi-axis sensor and initial measurement platform.

1:04Essentially, the motion sensors are used to monitor and control movement and also do things like navigation. In this area, they represent data that will be referred to as a digital biomarker. marker. And these digital biomarkers can consist of like a gate or arm swing or neck movements. Some of the other MEMS technology that we have includes pressure sensor, magnetometer, otherwise known as a digital compass. MEMS technology has been around as originally military technology was used for guidance systems for all types of things, including rockets. And back in between 2007, 2010, some companies, including Bosch, STMicro, TDK, InvenSense, developed the first consumer grade devices that eventually went into game controllers and smartphones.

2:05Apple adopted the first gyro for smartphones. That's a great entry point, by the way, because a lot of our listeners aren't going to be experts at MEMS or accelerometers. But if you say like, you know, it was used in a Wii game controller immediately, okay, how do you translate that into a healthcare device? Sure. The biomechanics labs at universities were very interested in this technology to measure human motion. Prior to that, and they still do, use camera-based systems for motion. And so any lab that you go into, you will see these camera-based sensor systems that are very expensive. If you have an accurate enough inertial sensor, MEMS, you could start replacing those cameras with first hybrid and then straight to sensors only.

2:58So instead of going in a green room, you could take that data anywhere, like on the playing field or on the golf course. And of course, humans move and react differently in a real game situation, right? Or if they're a soldier, for example, you do soldier systems, the same thing. Yeah. Even though it was used for training, they also wanted to monitor the health of the subject, whether it's an athlete or a soldier. Going back to our experience, when we were making these consumer-grade devices, we developed the first smart mouthpiece where we put our accelerometer for the first time in a mouthpiece to be used to detect concussion and impact.

3:41We were hired by the Cleveland Clinic to build a mouthpiece for them prototype, but the MEMS sensor enabled the whole system. Okay. So you add software to that sensor and you develop a way to monitor concussions and impacts and swinging around to determine whether or not you were concussed. And we looked at the technology that was developed 30 years prior. We looked at all the projects that were funded by the government, exotic technologies to try to get very high performance, which would be the tactical to navigation grade. So this is like kind of mil-spec, military grade, built in a big can that were on rockets.

4:26So we said, well, if we could make chip-sized IMUs at a commercial foundry where they do really large volume, no one's ever done that. So how do we get there? How can we make it at the wafer level, encapsulate all six axes, and actually get to navigation grade, which is 80 to 100 times more accurate than consumer grade? What does that level of accuracy really get you in life sciences in med tech? Well, life science and medtech, one thing that they complain about almost right off the bat is that, well, we could develop a step counter really easily with these consumer grade sensors, which is great because we didn't have that before.

5:10But if we really want to measure accurately, gate, and start to predict, do predictive analytics, we need more accuracy. We spent five and a half years developing a process at the commercial foundry, not at a university lab. We also were following a lot of patents because it's a completely new area. That's a good encapsulation of the technology, the product. Let's switch gears for a little bit and talk about applications in health tech, med device, life science. I understand that you started a fund. You have a real heart to figure out how to get this into market, get this into the hands of other companies.

5:46So tell me about some of the applications that you would love to support, kind of some of the ways you could see this being used outside of what you are building at your company. Working with subject matter experts in any field is really important for us because then we develop new approaches to the design of the MEMS to have a custom device for that particular application. So if I want to do navigation guided surgery, I need to interact with those subject matter experts, those surgeons that have been working for many, many years and have a two to fill on how to do this. So the platforms for health care kind of start off with like a wearable, but a medical grade wearable.

6:31So what does that mean? They'll take off the shelf consumer grade sensors and build a software layer on top of it. and that would be their product. And they would say, we're sensor agnostic. So you can have 10 different companies doing wearable using the same commodity off the shelf sensors. You're trying to make the commodity sensor better. You're trying to make up for its lack of accuracy by using the software. So we have a slogan that we use a lot and you can't software your way to a better sensor. If I use a navigation grade sensor, that's$10 ,000 and I build my software based on this high accuracy data, the end result is going to be objectively more accurate.

7:19So if 10 companies are developing a wearable and one of them is using a navigation grade sensor, their software combination is going to be superior to everybody else's. Now I'm going to play the devil's advocate for one second. And in that, there are certain tools in health where accuracy is not the differentiator. So it's like if my step counter is off by a few steps, I'm not that sensitive to that type of accuracy. And even if my blood pressure, it's like there are things that could be maybe a half a degree off and I'm still getting the right sort of life, wellness sort of data. So where is it the most important to have the absolute best to pay for the top?

8:05Well, yeah, so that is the standard response to people developing the software. And what you also hear from the suppliers, these consumer-grade devices, nobody cares about the extra accuracy, which is just false. And so when you have more accuracy, you have better predictive analytics and better software. And the most advanced software now is AI machine learning. So it's always garbage in, garbage out. Less accurate data means worse results. And so if you look at a, we'll go back again to wearable. Wearable is collecting data. It's using AI machine learning for the optimum outcome, which is anomaly detection, right?

8:54You're looking for differences. So you have more accurate data. You're going to pick out more anomalies. You're going to be more accurate. Your predictive analytics are going to be better. Your AI machine learning is going to be better. When you take any wearable IoT device, edge device, and you want accurate data, the most expensive high accuracy data you're going to get off an edge device, off IoT device, a wearable, is the inertial data. Because that accurate inertial data is extremely expensive. It is$3 ,000 to$8 ,000 per access. So a ring laser gyro, fiber optic gyro that goes into a missile or a drone is expensive for a reason.

9:39And you'll never use that to put on a wearable, right? So I can't take 13 Honeywell ring laser gyros and put them on a motion capture suit. First of all, they're too big and too heavy. And if you do that, you're going to have several hundred thousand dollars worth of mocap. So question, would higher accuracy be preferred in health care or medical space? We haven't had anyone to say the opposite. They will make an argument that you don't need more accuracy because they don't have it. Got it. That makes sense. Tell me more about this fund that you started and what you're hoping to invest in. We're looking to invest in companies that we may partner with, license to.

10:30So we have a combination of patents, trade secrets, know-how, even copyrights for our semiconductor design. So we have multiple layers of IP protection. I used to work in investment banking a long time ago. And before I got involved in startups and developing new technology, I kind of left that. left that world to live off the land and get operations experience, but also work with engineers for a long time. So it sounds like you're pretty open in terms of the applications. You want it to be, you know, novel. You want to, you know, own the IP, but otherwise, hey, if it's a good application for your sensors, you could get excited about it.

11:18Yeah, we don't need to own the IP. We want to help, if we invest in a company, we want to help them build their own IP mode as well. But when you take unique complementary IP along with ours together, then again, you create more modes. If you have really deep knowledge on the sensor side plus the software side, you're much better off. We're pretty open right now. We know maybe 30, 35 percent of the applications that we could actually develop. In other words, we're going to come across them as we try to figure out and answer questions related to like predictive analytics, for example. You know, what are the anomalies?

12:06We're looking for a combination of anomalies. and that's going to be where a lot of gold is. Can we figure out and predict an injury of an athlete by taking all his motion data and that motion data says, well, this person is tired. And when you're tired, you're much more likely to get a ACL injury. Yeah. Right. Or to get concussed. That's right. And then you take action before it happens. So, you know, with dementia and being part of the Alzheimer's moonshot program with you guys, and I've learned through personal experience, family members, the sooner you can detect dementia, for example, the much better off you are.

13:01You know, just preparing, you could actually change the outcome. You could stop cognitive decline by changing a few things, not just exercise, but diet, because you're aware that you're at the beginning, early stage. And if you could delay it long enough, you could delay it 15, 20, 25 years, and it's not going to matter. But man, does it mean a lot if, you know, you could avoid it for if you're 50 or 60 and you could avoid it for another 20, 30 years. What if you're 30 or 40? Right. So that that really it's a big deal. Yeah, it's a big deal. And it motivated us to, you know, we got to get the technology out to the right people, find the right partners, subject matter experts, because we certainly are not in so many different fields and areas.

13:55And if we team up with those people, we could probably do some really unique approaches. I think that's a beautiful note to end on. It's a call to action to folks who might be hearing this that you are open for business and want to support various endeavors with your sensors and with your technology. What's the best way for people to learn more about the fund or get in contact with MEI Micro? Well, if you're a member of Startup Health, we're part of the network. Does the fund have a, is there a way to learn more about the fund online or is that still to come? Yes. So next month, we're pretty sure we're going to light up two new websites.

14:40We're still working on it, including R &R, so R &R Ventures. We'll make sure we link to that in the description of this content. Awesome. Well, Lewis, I appreciate what you're working on. You know, there's been so many applications of this technology in the military and in entertainment and the Wii. But for you to focus on healthcare applications and say, look, if we get these sensors into the right devices, we can catch disease decades early and give people more life to live. That's not just a product, that's a mission. And so I appreciate you dedicating your work to that. Yes, absolutely. And anyone who has the same mindset, we'd love to work with them and at least discuss the possibility.

15:26So we're open to that. And we might come up with new ideas together. And we do have a way to also kind of invest. And we want to reach out to other investors, be part of their syndicate and help their project and vice versa. Excellent. All right, Lewis. Well, let's make that happen. And we appreciate getting the update today. Great. Thanks. Take care. And finally, an invitation to all of the extraordinary health transformers, whether you are a startup, someone building the future of health, or whether you are an investor, funder, or buyer of the solutions out there in the marketplace, this is an invitation to you to join the Startup Health member community.

16:16Learn more at startuphelp.com. Look forward to hopefully seeing you soon.