Brian Greene Presents NeuronSphere at LSI USA '23

Transcription

Brian Greene  0:05  

Good morning. I'm Brian Greene, CTO and co founder of Neuronsphere, talk about platform engineering for data. And we're gonna start talking about what is a platform engineering. We talk in med devices a lot about, you know, going after a platform play, we're going to create one lower layer that we can then expand and do lots of things with, we think that that approach is viable for data. And if you've listened to any of the presentations, it's been data, data data, right? Like it's a device. And then here's what we're going to do with the data. So let's talk about that a little bit. My co founder, Kevin, he's sitting right out there, Kevin and I met 10 years ago at Stryker, we've been building technology teams our whole career independently. And in 2020, the two of us and two other engineers started NeuronSphere, we close to $2 million friends and family seed round, got started building got started with some early design partners and medical devices. And by December 21, we had a soft launch of our neurons fear suite with a few customers, and Feb 23, we have three that are using the platform. And we'll talk about what they're using it for. So why did we, why did we build this and what got us started? You know, if you listen to the keynote this morning, you heard an abridged version of this. But as Kevin and I were talking in 2014 2016 2018, what you find is that this is the emerging model, not just in med devices, but we'll talk specifically about med devices, this has been the emerging model, we're going to create a medical device. And maybe there's no machine learning, there's no data in it in the beginning, maybe there's a little bit of a trained model. And this can come from all kinds of places, it could be a robot, it could be an EMR, there's some algorithms that are using bed sensors out there. We've got wearables, we've got chart data, we've got, you know, radiology images. So let's go get a little bit of data and a little model. And let's go get that into device and let's go get it in the field. And then let's let it run for a bit. And as it's running, we're gonna gather more information, maybe we're gonna get lucky, right. So maybe we've got a device like Fred was talking about this morning, where we can capture the input of the clinicians, we can really understand that interaction. But now we're going to take that we're going to create a new model. And we're going to create a new revenue stream and a new benefit stream for the patient, new benefits stream for the clinician. So we looked at Med devices, as we were looking around over and over and over. This is the multi year plan for a device manufacturer. So where else does this interact? Right? Like, where else are you going to use your data? We're going to talk about research and development. But what about the academics, usually, we immediately want to share data with academics. So if you start using NeuronSphere, we help you share your data without having to copy it. So there's some privacy built in and immediately getting some of those early algorithms under control. You move into quality and compliance, right? There's lots of compliance, the FDA is getting better and better at regulating machine learning. But you've got to have a system in place that can capture all of the evidence that you need. So there's that angle, then you get into manufacturing, more and more. Even the the disposables, or semi Reusables have some intelligence in them, they have serial numbers attached to them. So tying into not only internal manufacturing and testing, but your CRO, how do your test jigs that you're putting out at the CRO? How do they give data back so that you can see it from the beginning to the end, then we're going to get into support and maintenance. Right field support has a very different set of use cases than r&d, are trying to filter things. They're trying to figure out real problems versus false positives, which there are always lots of false positives. So then we're going to get over to the patient and the caregivers, and everybody's got a mobile app. Now we're seeing more and more of those lots of these presentations. There's, here's how this mobile app is going to show the caregiver and show the patient some distillation of the information that came from the device. But we're also going to feed some of that into the health system. And this can be monitoring disposables to keep the supply chain lean. This can be billing, you know, passing information through the EHR, this can be passing diagnostic information through to PAC systems, your device, producing this data and being able to offer that back into an integration of the health system closes a big loop. Clinical looking for new evidence looking for real field usage. How is the device performing? What are people doing? Is it meeting expectations? Is there a new indication we can create with the data that we have there coming along? Now we're gonna get into marketing and sales? Right? How are we creating a new digital product that the salesperson can sell? Do we have integration to the EHR that makes this easier for the hospital system to get reimbursement How is that closing that loop all the way to commercial. So if you listen to the data story, everybody's talking about the the r&d part, and maybe some of the machine learning. But really, the data from the device can power all of your processes all the way around. And it's more than more than connectivity at this point. So this is why we built NeuronSphere, we wanted to really be able to affect that. We look at small startups, medium sized startups, trying to hire large teams to go conquer this problem over and over again, we think there's a lot of this that can be broken into a single, consolidated platform. So we talk about an integrated suite with extensibility, right? Here's how you can buy something that gets you from one end to the other, you're not buying 16 little tools within your own sphere. What's the timeline? You know, I hear people talk about well, in a couple of years, we'll have the team and the platform put together, we'll be able to really do something with our data, right? I think that's irrational, we have lots of technology, it's far more advanced, you know, 60 days with an asterix, you know, sometimes you can see some early results a lot faster than that. And then staff, right, you shouldn't have to hire an army to get this kind of thing done. So we talked about a NeuronSphere knowledge engineer that's trained on that entire lifecycle, and can be trained on a single toolkit with a single set of conventions from one end to the other. So it's a workflow optimization for how are we getting that real value out of our data? So what is the opportunity look like for this? You know, for me, some of the evidence, the opportunity is the keynote this morning, the future of the operating room is digital. It's listening to all of the presentations, I think almost every one that I've heard has been, here's the device, here's the data, we're gonna gather, here's the, you know, here's the plan for creating new algorithms. Right? I know, I've heard a couple of times, I want to, you know, we're gonna go create an app store so that we can build multiple algorithms on top of the data that we have. So we think there's a huge opportunity here, if you go do the math on how many firms are out there, and what it really takes to go chase this dream yourself. We think there's at least a $2 billion market out there in just the top half, or the top 1000 med device firms. Primarily, we're going after high complexity devices. So we're gonna have to robots, Edge machine learning, digital health, wearables, remote patient monitoring, all of the different indications you're seeing with imaging and radiology. And, and once we, I think, get the firm foothold there, we expand into other regulated industries. Because if you listen, there's 10 Different drone companies out there that are telling the same story. Here's how we're gonna fly our drone. Here's how we're, you know, Spot the robot dog. Here's how we're going to take that data. Here's how we're going to bring it in. Here's how we're going to make a new algorithm and then we're gonna put it back on the device. So how do I make that loop faster? It's kind of been Kevin and I's mission since we got started. And we think there's a big opportunity. We're profitable at this point. But we're always interested in investors. Thank you much.