“Businesses in manufacturing and beyond face significant challenges today, whether their reshoring operations, skill shortages, labor shortages, or political strain, but today’s automation technologies can help solve these problems.”

In this episode of the Manufacturing Matters podcast, Darcy Bachert, CEO of Prolucid Technologies and Chairperson of the A3 Board of Directors joins TECH B2B Marketing’s Winn Hardin and Jimmy Carroll to take a high-level look at some of the biggest challenges in manufacturing today and how automation fits in. Bachert also dives into some of the pain points that companies in the nuclear and medtech fields are experiencing today, and how his company has made their jobs easier through automation. In addition, the conversation touches on the role of AI in manufacturing today, working with integrators and engineering design service companies as a software development company, and why young people should start gravitating toward manufacturing jobs.

Jimmy Carroll: [00:00:07] Hello, everybody. My name is Jimmy Carroll. I’m the vice president of operations at tech B2B marketing. And welcome to this episode of the Manufacturing Matters podcast, where we discuss the latest trends in technology shaping the manufacturing industry. Today, I have the pleasure of being joined by my friend and colleague, Winn Hardin. Hey Winn and Darcy Bachert, who is the founder and CEO of Prolucid. Darcy, thanks so much for taking the time. Really appreciate it.

 

Darcy Bachert: [00:00:31] Yeah, thanks for having me and excited to have a conversation here together.

 

Jimmy Carroll: [00:00:35] Absolutely. Yeah. So for those who don’t know, you’re pretty well known guy in the space. But for those who don’t know, tell us about lucid and what you guys do there.

 

Darcy Bachert: [00:00:44] Yeah. So long history. We’ve been in business for 17 years now. And over the first roughly half of that. So, you know, around around 8 or 9 years, we did a lot of work as a systems integrator in manufacturing and automation robotics inspection. So a lot of the things that you would traditionally see within this industry. And then as part of that, we actually got introduced to projects in both the nuclear as well as the medtech space. And over the next few years, we really transitioned, kind of taking some of the same automation, inspection and robotics and different solutions and apply them to problems in those spaces. And so now what does we really focus on software based development for medtech companies. So they’re looking at bringing new medical devices to market. A lot of these are imaging and diagnostics and devices that will be in the operating room. That’ll be in clinics that are really helping doctors and overall improve patients quality of life and and getting better solutions to them. And then on the nuclear side, similar a lot of automation that’s helping improve the reliability of these these these facilities to make sure they run safely. But more recently a lot, a lot of rebuilds and new builds. We’re bringing software and automation in to help run those projects faster and better.

 

Winn Hardin: [00:02:06] You know, your company has always made me the most excited, Darcy, because you’re one of you’re at the forefront basically of transitioning what’s been industrial technology outside the plant floor. I mean, there’s we’ve talked a lot about that and there’s a lot of folks who integrate that there. But you guys are actually developing the products and, and, and of course that we’ll talk more about it. But that means that there’s a whole lot more disciplines that you have to bring together. So I can’t wait to get deeper into into that discussion. But before we get into that, we should mention also that you are the chairperson of the Board of directors this year, I think. Congratulations. Is that correct? We got that one right.

 

Darcy Bachert: [00:02:37] Yes, I appreciate that. Thank you. It’s been part of a three, I think, going back seven, eight years now and then joined the imaging division and Imaging Board first and then moved into the chair role a couple years ago. They’re taking that on and then as part of the board of directors. So it’s it’s pretty exciting. You know what A3 is doing the automate conference, all the other shows. I know there’s the focus conference coming up in Seattle, which is an AI and imaging one. So a lot of work that they’re doing is is so impactful. And really I’m thrilled to be able to be part of it and help with that, help with their journey and what they’re trying to achieve 100%.

 

Winn Hardin: [00:03:17] And for those who don’t know, all fans of the show are going to know the A3 is the Association for Advancing Automation, and they’re North America’s leading standards, development and trade association, focusing on machine vision, motion control, artificial intelligence and robotics. So all the manufacturing technologies. So in your position, you’re kind of the spider at the center of the web. Darcy. So one of the first questions you want to ask you is right now what excites you the most about our industrial automation landscape marketplace?

 

Darcy Bachert: [00:03:44] So what excites me the most is just the the massive opportunity that we have. And to be honest, some pretty big problems that we need to solve and truly believe that automation is the way that we do it. So there’s a lot of reshoring that we’re trying to do. We have a skill shortage. We have labor shortages, and these are all areas where A3 can can fit in. We can help with training and certification and mentoring of those next generation of workers that are going to come in to more highly sophisticated roles, where we’re helping drive automation so we can have a huge, a huge part in that, making sure customers see the right solution providers, systems integrators, manufacturers, OEMs as part of that ecosystem. So huge opportunity, and it’s a pretty exciting time to be able to help be part of that solution and solving these big problems.

 

Winn Hardin: [00:04:41] Absolutely.

 

Jimmy Carroll: [00:04:43] Yeah. Wind kind of nailed it. Like you being at the center of all these technologies and, you know, in the overall space and and like I saw like the physical embodiment of that. I think I told you that the first time I met you, but like, I would always see you sort of in the middle of a group, and that’s in the center of that big hotel in Florida during the 83 business war. And I’m like, I know, I know who Darcy is. I’ve seen his presentations. I’ve got to meet this guy. Uh, and I’m glad that I did. I’m glad that you’re here today. So I do want to ask, like, you know, you mentioned some of these problems. So what are, you know, both within, you know, nuclear and med tech and and beyond. I know you talked to you have a lot of friends in the industry. What are what are some pain points. What are some of the main pain points, I guess, that manufacturers are facing today, including a labor shortage like you mentioned?

 

Darcy Bachert: [00:05:26] Yeah. So that’s a great question. And I can kind of dive a bit more into what we’re seeing in both the medtech and the nuclear space, because there’s a lot of a lot of those themes are fairly they cross over these different industries. So on the medtech as an example, there’s a there’s a lot a large shortage of radiologists. You know, these are the people that take a look at our x rays and EMRs and CTS. They read them and they diagnose them and say, here’s what, here’s what exists. And then they can pass that off to a physician or a surgeon who can actually take that and do a procedure. Uh, there’s, you know, just a shortage of doctors. There’s a shortage of all these people in general. And we can either train and hire and try and develop lots of them, or we make their lives way easier through automation. And we can do that through. There’s things like agentic AI where we can. That whole triage and intake process, we can automate a lot of that. There is taking these images and doing a lot of work on them so that by the time the radiologist sees it, their their job can just be to look at it and say yes, no and not have to go through, you know, a large processing workflow to get to the point of making a decision. So there’s all these opportunities to make all the professionals way more efficient, allow them to do what they do great and do much more of it and less of the bureaucracy and and input just to really get rid of that.

 

Darcy Bachert: [00:06:49] So it’s it’s not that dissimilar. You know, we want to bring manufacturing back. It has to be done effectively. Uh, we, we have to drive the efficiency in how we produce, how we inspect, how we, you know, get things out the door. So we’re doing the same thing on medical and then, you know, on nuclear, the history of just projects that would run over way over budget. I’m sure you’ve heard before, you know, you’re going to build a new facility or maintain it, and it cost overruns and things like that. They’re just not acceptable anymore. And so what we’re doing is we’re actually using software automation to improve time on critical path. Uh, when a reactor is down, we’re talking about hundreds of thousands of dollars an hour or millions of dollars a day. So shortening that time frame where they take them offline to inspect them using automation makes them more effective. It makes it better for the utility, for the taxpayer, the ratepayer, everyone. And then on these refurbs where, you know, every 30 to 50 years, you got to tear the thing down and really almost rebuild it from scratch to make sure it can run for another 30 to 50 years. If that takes a year versus automation, allowing that to be done in eight months, that’s a significant cost savings and impact. And so it’s some of those same themes. It’s shortage of skilled laborers. It’s also driving efficiency. And that’s ultimately what automation is all about is making making these things better faster. And ultimately the end. The end patient or user is the one that really benefits.

 

Winn Hardin: [00:08:21] Right on. You know, Darcy, when you and I first met maybe ten, 12 years ago, I’m not sure, but I thought of it kind of initially as like an embedded systems house. You know, you brought together all kinds of different disciplines and then, you know, learned, you know, as Maura would, you know, we would talk about you’re doing product development. You’re almost like a specialist engineering house, right. Who’s helping largest OEMs out there to create consumer product goods in many cases that are leveraging this technology. And now, you know, we’re talking about a focus in medical and we’re talking about a focus on nuclear. Can you talk to us a little bit about what makes what you do different from those. Your industrial system integrator roots. Right. Do you have to approach projects differently when you’re developing these? Is there more technologies you bring together? How are they different? How are they the same?

 

Darcy Bachert: [00:09:07] So there are some differences that are that are very clear that I can share. I guess at the core of it, being able to run projects effectively is the most important and the most challenging thing. Customers have an idea of what they’re trying to build and why, but it’s not always well understood. They don’t understand the limitations. They don’t always understand exactly how it’s going to fit into the end user workflow. So the biggest thing that we had to do in going from more traditional systems integration, where you would get a, an RFP and a machine spec, and it was pretty clear what you had to build. There’s a lot more ambiguity into what exactly needs to happen. So we spend a lot of time up front really diving into what is the problem that you’re solving? Who is the end user? How will this new product or technology fit into their workflow? Because if it does not make their lives easier and better, we can build the most beautiful thing ever and it’ll never get adopted. So we spend a lot more time up front really challenging that. We’ll talk to nurses, we’ll talk to doctors, we’ll talk to, you know, people that are managing the nuclear facilities and really just understand what they do, what’s their day to day activities and then bring that back. And once you’ve got that project clarity, you know what you need to build and why. Then you know then that execution is a lot easier to do. And that’s the I guess the other difference with that, the regulatory side, the quality management system, cybersecurity requirements, there is a lot more. Additional overhead or things that you have to get right, because at the end you can’t just release it and sign off. In a way it goes. It has to go through regulatory bodies. Who are the ones that will say yes or no, this is this is good to go.

 

Winn Hardin: [00:10:53] Is that is that something that’s, uh, particular to mission critical systems like medical and nuclear, which, I mean, we’re, you know, the cost of failure can be even higher than dollar amounts, right? Or is that is that something that’s more common to an embedded system developer or, um, what’s the proper way? How do you refer to Prolucid right now? Darcy. Tell me what’s what’s your catchphrase? Catchphrase.

 

Darcy Bachert: [00:11:18] So we are a full service software development firm. That’s what we would describe ourselves as. And so whether we’re working in nuclear or in medical, you know, we do tweak a little bit what we do for each one. But we’re a full service software development firm.

 

Winn Hardin: [00:11:32] Okay. But you still have hardware components to what you guys are. I mean, in terms of specking out and hardware and software. Hand in hand. Right?

 

Darcy Bachert: [00:11:39] Absolutely. Almost everything that we do is software and hardware working together. The piece that we really drive value on is the software side of it. But we’re having software that runs with SCADA systems, with PLCs, with robotics applications, with embedded hardware. Almost always there is a hardware component that’s very central to it. And so we have to understand that piece very, very well. But what we’re doing is really doing the software layer on top of it to the best that we can.

 

Winn Hardin: [00:12:06] You know. And that makes sense. We’re seeing that in all industries right. Software is a differentiator. I mean, the hardware in many cases, you know, there’s a lot of cutting edge stuff, whether it’s, you know, it’s hyperspectral, advanced 3D Imaging system, lidar systems where hardware is still evolving and it always will be. But the computational platforms are so strong now that, you know, the differentiator. And the secret sauce is often in the in the software environment. So, um.

 

Darcy Bachert: [00:12:28] I would agree, just because we’re getting to the limits of physics on so many things with ultrasound, with imaging in general, you approach the limits of physics for what it can do. And so better resolution doesn’t always matter. But being able to take that image and identify cancer, identify, you know, different, um, things that you’re looking for in a more automated manner. That’s where a lot of the opportunity is now.

 

Winn Hardin: [00:12:53] Right, right. Absolutely. And before we move on, just one other comment. You know, for for 20 years that we’ve been in this industry talking about this industry, one of the biggest challenges for any systems that decide whether a system succeeded or failed usually depended on the quality of the specification, the scope of work that was initially created, you know. And now when we’re talking to folks like you, we’re talking to cancer over composable about developing intelligent AI agents and everything. There’s such a focus on the contextual components. I mean, now you’re seeing system integrators who are like, I have to sit down with the folks who are already doing this job. You know, we’re we’re trying to mimic more, uh, human, uh, A level knowledge making or process management or whatever it is. So we really need to be looking beyond just, you know, moving something from A to B. Um, and I just wanted to emphasize that seems such a critical and I would suspect that even in smaller deployments, um, folks that take a little bit more time up front to really develop a really comprehensive specification and requirements document, including the human components, uh, even at the smaller system level, are probably going to have higher levels of success and customer retention and everything. Is that a fair statement?

 

Darcy Bachert: [00:14:06] It is. Absolutely. It’s it’s ultimately product market fit means you’ve built something that the end user really wants, and it makes their lives better or faster or it has a real positive impact. And so if we can really understand that and make sure that gets right into the core spec of what we’re building. So the product requirements, then the software requirements, if we know that those being Achieved means they’re going to get product market fit. I think for any company out there, you know, whether you build a widget, a complex device or a custom solution, that product market fit is everything and getting it right is incredibly valuable.

 

Winn Hardin: [00:14:44] One more follow up question, Jimmy. And I swear to God, I’ll shut up after this one for a little bit. But do you find do you find that customers are more willing? I mean, I’m so used to customers coming to us, you know, even in our little marketing technology space where we live and breathe and they just expect you to know the answer, like, here’s the problem I have. Just tell me what the answer is. You know, I mean, and and the and the information gathering, the discovery and looking at all the different systems and components and everything. Is it getting easier for for customers to understand? Maybe there’s a baseline knowledge. So now they understand this isn’t something where I can just take anything out of a box, put it over my line or whatever, you know, fire up a software program, hit a few boxes and be done. I mean, are they starting to understand we’ve got to do this due diligence or is that still a hard sell?

 

Darcy Bachert: [00:15:27] We’re Are. We’re finding people to be a lot more receptive to it. And I think it’s also because we’ve gotten better at messaging why it’s necessary. Right? What I’ve always said is that when we look to work with someone, if someone asks really good, intelligent questions, it shows that they’re engaged. And when they ask those good questions, it makes you think. And then that really just continues. The momentum goes I think. So we always start with is trying to ask really good questions, right. Challenge some of the assumptions, but also show we’re doing this because we want this is what we want to get to. We want to work with a company that’s going to be very successful. So if we can ask some good questions, challenge assumptions, and then collectively build a much better solution, that’s a win win. Now we’re going to have a customer for life. And I think in what you guys do, it’s the same thing. You know the better the questions you ask, uh, you know the better the engagement is, the better the response is. It’s just it’s where it all starts.

 

Winn Hardin: [00:16:26] 100%. Yeah. Can’t execute without a good plan, that’s for sure. But it’s good to see the marketplace is starting to understand a little bit more. And I agree with you asking intelligent questions and then backing that up with POCs and data points to prove, hey, you know, if we do it this way, if you have you considered this component. Here’s where we’ve seen people succeed or fail in that space.

 

Darcy Bachert: [00:16:45] So if someone doesn’t have an answer, that’s okay. Let’s go find that answer. Let’s do the POC. Let’s talk to the people that we haven’t talked to. So I think sometimes there’s this uncomfortable saying, I don’t know. But when we get a customer and say, I don’t know. Okay, great, let’s go figure this out. Because once we figure it out now, our chance of success is way higher.

 

Winn Hardin: [00:17:04] Yeah. All right, Jimmy, I’m going to shut up for the next 3 or 4.

 

Jimmy Carroll: [00:17:08] You know, I’m curious here because I’m. You know, you guys are a software development firm. Um, what’s the process look like when you’re working with a company that’s developing a new system in terms of working with the people that are installing the hardware or whatever it might be like, how do you how do you partner with other integrators? And like where does that begin?

 

Darcy Bachert: [00:17:29] So we are you know, a lot of these projects, whether it’s an integrator or a so on the medical side, they’re more like contract design houses where they’re designing electronics and hardware. There’s always an area where, you know, the systems need to really join together and work. So if at all possible, right at that, that early discovery phase, let’s understand what they’re doing, what we’re doing. Let’s challenge all those assumptions. Create a really clear interface for, you know, what they where they stop and where we start. A lot of times on the medical side, someone that’s designing an embedded device, maybe they’re doing the firmware, but we’re doing the the mobile, the cloud, the, the desktop application. So we want to have a really clearly defined API for how we’re going to talk back and forth on the nuclear side. We’ll be talking to SCADA systems and robots and PLCs and collecting sensors. So again let’s understand all those interfaces upfront. Let’s do demonstration points. So every single week we make sure to demo what we’ve done to the the stakeholders. Here’s what we’ve done. Here’s what we plan on doing next. Here’s what’s blocking us. Because one of the biggest mistakes you can do, and I think in any business is gather the requirements, go away for three months and then come back and say, here’s what you asked for because you’re going to get it wrong. So those frequent demo points, and then as soon as we can start testing with the data, with the robot, get our software talking together, we do that. So it’s just the old waterfall way was you did your requirements, you did your development, you did this. And that was kind of your only touch points. Now, it’s like I said, every single week and sometimes more frequently demo get feedback, exercise that. And then when you get to the end, the whole mythical that software is 90% done when it’s really, you know, halfway there. It just does not happen. But it takes a lot of work every single, every single touch point to make sure that doesn’t happen.

 

Jimmy Carroll: [00:19:25] Yeah. Fair enough. So I know that. I know that, you know, I’ll preface this by saying, I know that you’re bound by the tightest of tight NDAs and you can’t talk about very much, but but in terms of like what that looks like in a real world example. Um, do you have any in either nuclear or medtech or both? Ideally, both. Maybe we’ll start with nuclear that you could talk about that you know, from start to finish what that look like and what you guys did.

 

Darcy Bachert: [00:19:50] Yeah, absolutely. So the this is one of the the videos that’s available. It talks about some of the work we’ve done with Bruce Power as part of their major component, um, refurb that they’re doing. So essentially they have eight reactors that over the last few years and the next few years, they’re completely rebuilding one after the other. And so what this project was doing is helping with automation of one of the steps in that process, which is called the inspect. So after they remove a lot of components. They have to go and inspect everything that’s been done before. They then go and install the brand new parts. And so as part of this project, it was us working with them. But there’s also systems integrators doing different pieces. So multiple partners that all had to work together. And we really followed that process of up front. Let’s really understand what the requirements are and the requirements actually changed over the course of this because we were able to expand scope and accomplish more. And that’s okay. But we had a really good baseline. We knew what the hardware and integrators were doing, what the end customer was doing, what our part was, have those frequent demonstrations. And then, well, before we get to the point of being on call it on channel or right on site, when you get on site, you’re now on critical path.

 

Darcy Bachert: [00:21:04] Things have to work. We were able to have a series of mock ups that were done, ranging from a very small mock up at the integrator location, where we can go and test our software and their hardware, all working together all the way to Bruce Power. They have a staging area where they have a real, you know, full size mock up of what it’s going to look like when you’re actually inside the reactor. And there, again, we could go through an exercise, all the software, software and hardware working together to get to the point where, you know, then once you transfer it across, yes, there’s going to be some setup and install, but things are just going to work. So that’s a that’s a very typical process that you would do. And you know, because that that on channel when you’re on Critical Path is so valuable. Getting it right before you get there becomes even more and more important. Uh, sometimes on the manufacturing side, you will have a few weeks or maybe a month to get a line up and running. We don’t have that luxury here when when we get there, it has to work.

 

Winn Hardin: [00:22:04] No pressure.

 

Darcy Bachert: [00:22:06] No pressure at all. Funny enough, when we were doing the install, one of the team members in kind of a joking way is like, you know, you’re going to be on critical path in a couple of days. No pressure. If you get it right, no one will know you were there. If you get it wrong, everyone will.

 

Winn Hardin: [00:22:19] So that’s the story of life right there. Except success also breeds success. So that’s a lovely thing too. How long did that particular project take you from beginning to end? There’s a lot of different components, a lot of different partners collaborating in on it. So I mean, it’s not like you’re in control of the full development schedule.

 

Darcy Bachert: [00:22:36] But yeah. So, you know, once we got fully there’s a lot of upfront work that happens on gathering the requirements and and making sure everyone’s aligned on what we’re building. And that that tends to take a bit more time to do. But once we really got kicked off on the project, more or less just less than a year from being starting that to doing the testing to actually being, um, uh, being on, on, on channel, doing, doing the work.

 

Winn Hardin: [00:23:03] And how long about did the discovery phase take? I’m just curious because, again, I think it might get back to the the point we were talking about earlier, the importance of knowing what you want and what you need to accomplish, you know, before you start.

 

Darcy Bachert: [00:23:14] You know, that whole phase was a couple of months as well.

 

Winn Hardin: [00:23:17] Okay.

 

Darcy Bachert: [00:23:17] And typically on the on the medical side, it’s a little a little bit better than that. We would typically say you’re going to need 4 to 6 weeks to do discovery and really understand it. And that’s because we have to go talk to stakeholders and get their input. And that would then translate into maybe a 4 to 6 month long project where at the end of that, we’re now ready to have a version that’s for clinical trials or potentially all the way for FDA and, and commercial use. So it’s it is a bit more condensed there. And I think we’ve really got it perfected. But on the nuclear side, some of the projects are larger, there are more stakeholders. And that just takes a bit more time.

 

Winn Hardin: [00:23:55] Gotcha. So I mean, from those two examples, it sounds like somewhere between 10 to 15% of project schedule is done in discovery, you know?

 

Darcy Bachert: [00:24:04] Absolutely, absolutely. And if there are unknowns and we have to do POCs, that becomes even more tilted because sometimes we’re pushing the boundaries of technology that exists, and PLCs are needed, and that can easily be another 1,015% of the overall schedule, if that’s needed.

 

Winn Hardin: [00:24:20] But for a multi-million dollar project, increasing the chances of success from, you know, 78% to 98% is is worth every dollar. Not to mention, when you consider the downtime cost that some of these industries that you’ve been talking about.

 

Darcy Bachert: [00:24:34] It is. Absolutely. And on the medical side, a lot of these are earlier stage companies. And it’s not just costing them money, it’s if they get it wrong, the company won’t exist.

 

Winn Hardin: [00:24:42] Right?

 

Darcy Bachert: [00:24:43] And so it’s it’s survival. So you’re you’re a pretty core part of that, that solution, which is also very rewarding when you when at the end it works. It’s it’s pretty cool to see.

 

Winn Hardin: [00:24:54] Well man, I’d like to have another beer conversation where we just talk about those conversations with those customers in the startup mode. Uh, you know, we really want this super complex system and, uh, we really can’t quite afford it. So we were thinking about, you know, how do you feel about swapping out some equity? I don’t know, I’m sorry, but that’s just. I can see that conversation. I’ve had a lot of start up conversations in my days. People bring good ideas to you. Um, some of those deals get structured interestingly.

 

Darcy Bachert: [00:25:18] Yeah. And, you know, for us, we typically we do really focus on the companies that have a strong team, a strong technology, because we know they’ll be able to get funded and upfront. We’ll invest a bit of of time in that discovery process to help give them a better plan. Yeah. Knowing that once they get funded, they’ll come to us. So I think there is like we’re invested in the ecosystem. But, you know, if you really pick the winners and do a great job, then I think that investment pays off.

 

Winn Hardin: [00:25:47] See, Jimmy, we’re doing what Darcy does. So that’s makes me feel better. Yeah.

 

Jimmy Carroll: [00:25:53] Absolutely. Oh, you know, as far as picking the winners, do you have any examples in the med tech side, you’re you’re able to talk about even if it’s in broad strokes.

 

Darcy Bachert: [00:26:01] Yeah, absolutely. There’s a couple in particular that I mean there’s lots that we have that I think are going to do Fantastic. Uh, there’s one company called Opticon. They’re based in Australia, and they have a device that’s an imaging based device that they’re using to look for cancerous cells. So it’s, uh, it’s actually look at like a think of, like a, a large optical stack that’s been miniaturized to be the size of a pen. And so as part of a large, you know, as part of a surgery right now, what they would have to do is take a biopsy, send it to a lab, or have another machine that sits there that looks at it to see whether or not they got all the margins of cancer as they’re removing a tumor. Whereas now what we can do is use that imaging device to do the imaging. The part that we help with, though, is be able to then stream that to a pathologist. That could be anywhere in the world in real time, and they can then say yes or no. So think about rural or smaller hospital centers where maybe they don’t have a pathology lab or someone that’s trained, or you just don’t want to have extra people in the O.R. so being able to solve that problem, just takes the value of that product and amplifies it.

 

Darcy Bachert: [00:27:09] So we’re pretty pretty excited for them. And there’s another company actually shared this one at the business forum a couple years ago called Vita. And they are looking at triple A. So abdominal uh abdominal aortic aneurysms. And right now the current standard is they’ll do imaging to measure the size of the aneurysm. But small aneurysms can rupture and large ones can be fined for life. And so it’s a pretty poor way of determining whether or not intervention is required. What they’ve done is built. And we helped build this product with um, take time series MRI images. So as the heartbeat happens to see the whole expansion and contraction. Run that through a complex processing pipeline and then generate a strain map. And based on that strain map they can see whether or not it’s weak. And that’s what actually is the determining factor of whether or not intervention is required. Quiet. So there’s some really these applications that are taking the standard of care and just transforming it, making it way better. And ultimately, you know, I still consider myself young, but, you know, we’re all going to need these types of things. And the better diagnostics that we can have, the better treatment I think we all benefit from. So a few of those examples, it’s just it’s pretty awesome to be part of that journey.

 

Jimmy Carroll: [00:28:27] Oh yeah. I mean, that’s what I was just thinking about. Like, I think in a lot of ways, most people who, you know, in their profession, whether it’s like, you know, six degrees of Kevin Bacon wise, they can find some way to say what I’m doing is, is for the greater good. But you’re kind of like directly attached to that. I mean, it’s got to be fairly, intrinsically rewarding for you and your team.

 

Darcy Bachert: [00:28:48] It was even after that, that presentation, I had a couple people come up and say, I actually have triple A, and I did not realize this technology existed. When can I, you know, when can I get this? When can I get a scan or, you know, so it’s when that happens, it’s, uh, that’s pretty cool. And I always tell them that, you know, we can’t we can’t give you the scan itself. And it does take time, unfortunately, for these devices to get to market, it’s not a fast process. But yeah, the technology and the solutions are coming and we will all benefit from it.

 

Jimmy Carroll: [00:29:17] It’s very.

 

Winn Hardin: [00:29:18] Cool. Any kind of diagnostic system that doesn’t involve a scalpel, it makes me very, very happy. So yeah, you know, or having to stay open on the table for an extra hour or two while you’re doing the pathology check on the surrounding tissue. I mean, that, you know, you just don’t. And when we think about, you know, ballooning medical costs and everything, these are the types of systems that are going to make surgeries faster and cheaper, avoid a whole lot of unnecessary stuff. So, I mean, it’s not just avoiding human suffering, right? I mean, it’s actually helping societal potential issues. Uh, so it’s, you know, the benefits are many fold. I just it’s so cool.

 

Darcy Bachert: [00:29:53] Absolutely. And even the robotics, robotics adoption is really ramping up in um, in procedures as well. You’re seeing a lot more of it. So there’s some big players that are really dominating it. But you know, it’s it’s not to the point now of, you know, you just push a button and away it goes. A lot of them are more you know, the the surgeon is still operating a pen or a device, but they can be more precise in doing it. Yeah. There’s less complications afterwards because it’s more minimally invasive. So a lot of exciting advancements there would be interesting to see ten, 15, 20 years from now how automated it does get, because there still is that human element to it. But yeah, it’s exciting to see.

 

Winn Hardin: [00:30:36] Human enhancement right at, you know, at the medical level and many other places. Um, and the whole service robotics thing is going to be so exciting over the next 20 years anyway, so I can’t wait to see where that goes. Um, so we’ve talked about medical, we talked about nuclear. Um, and you’ve mentioned once or twice, I think, incorporating some artificial intelligence or deep learning models into some of your products. So can you tell us a little bit about there’s there’s always the ongoing debate. Debate is is AI overhyped or underhyped. You know, but you’re out there building things that are going into the marketplace. So what is your take on AI?

 

Darcy Bachert: [00:31:08] So we’ve we talked quite a lot about this with customers because early on yeah, we want AI to be part of it. And I think it’s more about company valuation or telling the story that actually is it the right solution for the job or for the problem. So we look at AI as can the traditional methods do it because if so, then it’s going to be easier to implement, easier to test, easier to control, easier to get through regulatory as well. And and then if not, then what’s the simplest way that we can do it with with AI? Is there an existing model out there that we can bring in that’s kind of pre-trained building the right data sets around it, the right infrastructure to support it. So it is a bit of a progression. There are cases where it is highly useful. It is. It really is the only way of doing it. And then now we’re getting more into seeing some of the Agentic AI being brought in to try and help with that triage process that happens up front. There’s a lot of intake questions that based on that, you know, whether or not someone’s got a maybe even what type of condition they have all the way to, are they a good candidate for surgery? So that one’s a little bit earlier right now. But you know, we always say is it’s if it’s not needed, let’s do it the simple way. If it is the solution, let’s use it the right the right way. And then a lot of restrictions, you know, it’s not adaptive AI. It is really locked down. You can constantly retrain it, but you have to have design controls around it. I think that’s important because if it was self kind of self-learning and training, uh, and that’s being used in critical mission critical operations or procedures and it goes AWOL. That’s that’s different than a chatbot becoming offensive. That’s going to hurt someone.

 

Winn Hardin: [00:32:59] You laugh because this is a serious topic, but that’s a timely comment. I’m just not going to touch it. Um, but but like I was wondering what when you put AI or a DL model into something, especially in medical, I mean, you’ve got FDR 2.11, you know, anything changes the entire process to be documented. Re verified. I can’t even imagine what requirements would be for the nuclear industry. I mean, are they even open to leveraging models? Or as long as you lock down the training model, it can go through certifications and be accepted. Can you talk a little bit about in highly regulated industries, what’s the acceptance.

 

Darcy Bachert: [00:33:33] On the the medical side is way further along right now with understanding it. And there are lots of just software as a medical device that is really an AI model with a wrapper around it that are on market and they’re making a significant impact. So, uh, there’s also very clear guidelines on the FDA and other regulatory bodies on how to do it. So, you know, just like anything else, the model itself may be a little bit harder to explain in terms of how the training works. But you have a model, you have a controlled data set that’s used for training. You have an independent controlled data set that’s used for verification. You have to have different subject matter experts labeling the data to eliminate bias. There’s a lot of rules around how to collect the data, how to label it, how to remove bias and to make sure there’s control over it. And then once that model is is built and tested, it is version controlled and locked. If you now have 10,000 more images that you want to use to make the model 5% more, you know better, you can update that. Now. You just have to run through that same that same control process, right? If you do all of a sudden change the indication. So instead of just saying this model is going to detect abnormal density of tissue, and now you’re going to say we’re going to detect a specific type of cancer. That would then require a whole brand new regulatory pathway. But for the most part, it’s very well defined what the process is to have it included as part of your product.

 

Winn Hardin: [00:35:00] Interesting. Would you say that medical is one of the vertical industries leading AI in terms of understanding certification processes, standards.

 

Darcy Bachert: [00:35:12] From a regulatory and certification process? We see them really leading the way based on the industries that we work within. Uh, you know, definitely on the consumer direct to consumer type of devices, obviously, that’s where a lot of the initial development is happening and that’s going so much faster. But as those as that research happens and those models get open sourced and become available, we can now pull them in and take advantage of it within medtech to solve bigger problems than we could have with, you know, the previous way you hire a team of researchers and PhDs to build a model from scratch, and that’s just not tenable for a lot of these companies.

 

Winn Hardin: [00:35:50] Yeah. Interesting. I guess, like industrial applications could maybe learn something from that. But usually what it comes down to in those is just the, the dearth of of, you know, enough defect samples, enough enough enough images or whatever the objects are to train instances to be able to train your model. Um, especially depending on how many classes of, of issues you’re trying to identify. Um, whereas medical, they’re just going to be like, well, if you ain’t got the data set, you ain’t moving forward. So work until you get it.

 

Darcy Bachert: [00:36:21] Yeah. That’s the one constraint that medical would have is control of the data set. And, you know, the FDA as an example, you have to collect data from the same demographic that your device will be used on. So if you’re targeting females within an age of 30 to to 50 for your device, that’s who you have to collect the data from. Whereas, you know, on the industrial side, not as many rules around that. And that also means in some cases development can go faster there. You see a lot of the imaging companies and robotics companies more and more. They have tools where either you can use their built in things, or you can bring open source models directly into their environment. So it’s the pace of innovation and adoption of AI within manufacturing as well as really accelerated in the last few years.

 

Jimmy Carroll: [00:37:09] Darcy, we’ve covered a lot today and what I want to be respectful of your time. Uh, is there anything else we haven’t talked about issues that you feel are real important that, you know you want to get out there?

 

Winn Hardin: [00:37:23] I think we let it go, man.

 

Darcy Bachert: [00:37:27] I think we can go on on these topics for a long time. It’s obviously really exciting to talk about. And, uh, I think maybe if there’s anything for for anyone that’s, that’s young, that’s looking at entering the workforce, you know, in a lot of ways we all want to go for the the high tech job because it looks cool. But I think within manufacturing, being able to now work on these solutions, you can get the both of best worlds. And so I think looking at the trades, looking at manufacturing, this is a going to be a huge opportunity and need. There are going to be the jobs won’t be, as you know, dark, dull and dangerous as they were in the past. There’s a lot of technology working together with building and making things. And you know, for me that’s always been rewarding to help build things. So I think just encourage anyone that’s looking at a career path. There are some pretty cool things that you can do in manufacturing in these other industries.

 

Jimmy Carroll: [00:38:19] I love that, and I appreciate you mentioning it.

 

Winn Hardin: [00:38:23] You know, it’s an excellent point. I mean, back in the day, if you were working, if you were a welder or something like that, I mean, for the most part, the welding machine, the the, you know, the head and everything, it is what it was. But now if you’re going into manufacturing, as the machines get smarter, the human actually has the opportunity to help shape the machine and retasked and reconfigure and optimize and everything else. So it’s not just being in the line going, getting getting, getting you know. I mean it’s manufacturing jobs are just completely different nowadays and even more so going forward. So.

 

Jimmy Carroll: [00:38:55] Well Dorothy, thanks so much. If people want to learn more I think it’s Prolucid. Is that right?

 

Darcy Bachert: [00:39:00] Yes. Prolucid or I’m on LinkedIn. Uh, feel free to reach out my email. Darcy I’d love to talk about these things. So anyone’s interest. Yeah. Please. Please contact.

 

Jimmy Carroll: [00:39:12] Yeah. And we’d be we’d be more than happy to pass along any questions that anybody has. You can reach out to us at manufacturing Matters.com. Uh, and if you’re interested in being on the show or, you know, sending comments, questions or anything to Darcy or us, please feel free to do so. Otherwise, thanks everyone for tuning in and we appreciate it.

 

Darcy Bachert: [00:39:31] Thanks everyone.