The tech behind patient-first transformation at USME

As a medical equipment rental company that rents, sells, and manages movable medical devices, including infusion pumps, monitors, ventilators, and incubators, USME’s mission is simple in definition, but highly sophisticated in practice.

“Our job is to deliver the right equipment to the right place at the right time,” says CIO Antonio Marin. “When you look at the community we serve, the last part of the supply chain is a patient in need. So we need to make sure all our technology, processes, and everything we do has a patient in mind. After all, they call us because they need lifesaving equipment, not because it’s a beautiful day.”

A particularly vital application of technology for Marin and his team has been directed to revamping the company’s inventory and equipment management, and field services.

“We did a lot of automation behind the scenes,” he says. “Knowing your inventory, knowing what parts you need to fix, and tracking the lifecycles of inventory is all now very automated, well managed, and fully visible across the organization. It’s about humans making critical decisions, not doing paperwork.”

But with that added efficiency comes some risk. And when lives are on the line in a highly regulated sector, vulnerabilities can surface with more tech that’s introduced. So some innovations can be more detrimental to the operations of a company or a hospital.

“We use encryption and different systems to overlay protection when it comes to personal identification data,” Marin says. “When you look at the cybersecurity chain, humans are still the weakest link.”

When talking about security, particular care needs to be taken in terms of knowing exactly where the team and equipment are at all times, and tracking performance across company and hospital staff, and hospital partners.

“As a person in IT and as an employee of the company, it’s very rewarding when we’re able to deliver lifesaving equipment so hospitals can succeed in helping patients,” he says.

Marin also discusses the importance tech and human synergy, prioritizing education in regard to cybersecurity, and the power of automating processes. Watch the full video below for more insights, and be sure to subscribe to the monthly Center Stage newsletter by clicking here.

On setting the right foundations: We’re in the middle of a major transformation. The company started with a homegrown system with phenomenal software, but as we’ve grown, it becomes more complicated to keep up with the rate of progress. So we decided to move to a SaaS platform and we have the first part of the project already complete. It’s been very successful and now we’re finishing the second part.

We can look not only at our business processes and refine them, but we think about embedding AI for faster and more accurate results. You have to have sound data and processes with AI. In one of my previous companies we used AI at the beginning when it was a buzzword and not really there. I learned a very important lesson then. You can fit the model, train it, and ask a specific question, but an unexpected answer might come back. So we went back to the old ways to analyze data and realized that the answer was right but the question was wrong.

I learned you have to be open to evaluate answers and understand where the real data is coming from, and the real sentiment on the data — the context of the information you’re working with.

On human involvement: There always has to be a human in the loop. That doesn’t mean we can’t speed the process for that human. There’s incredible things we’re doing today where an AI doesn’t have to be just gen AI. There are so many variances of AI and versions of what you can do with it. For instance, we’ve been able to automate the ordering process from a single click at a hospital nurse station to our branch operations where we get all the information we need to deliver lifesaving equipment.

In one hospital in particular, we delivered a full bed and mattress in less than 15 minutes. To put that in context, industry standards are normally between 12 and 24 hours. So in certain cases when we’re in proximity, we can be extremely fast because there’s no human interaction.

On AI and model training: We created a system called GoUSME Connect. It’s a combination of RPA, AI, and machine learning that can read a request generated by an electronic medical record system. So we’re agnostic of any EMR, and it reads information. And through machine learning, it reads the pattern of the request that transfers into an order, which ends up in one of our delivery locations.

That’s one part of how we can deliver equipment. We’re working hard to continue on predictive analytics and teaching the models because as a rental company, we have so much information about the true performance of medical equipment. Our goal in the next few months is to be able to predict equipment failures based on historical data.That’s the thing about medical equipment. It’s just a new computer. They have to go through preventive maintenance once a year, and every time they come back from a hospital, they go through review process.

So we always make sure equipment is patient ready. As we all know, though, equipment can fail. But if we can gather all the equipment we’ve rented in the last 23 years and start feeding those models with all that data, then we can be more predictive.

On logistics: One of the first things is to know your inventory, what equipment you have. And in the medical equipment rental business, it could be very seasonal. You have times where you have respiratory issues, then you get neonatal seasons. So what it allows us to do is look at our past rentals, and our inventory, and then start helping the equipment management team plan their production for the next month, week, or the next day. That’s a huge change in how we used to do things to what we can do now.

From the time of getting equipment prepared to being patient ready in the old days could be like getting a call, having a technician look for the piece of equipment, and then do all the necessary paperwork and testing. Every interaction was very manual. Now we know where it’s coming from and we prepare it. If parts for a piece of equipment are needed, the parts requisition is already requested. We know where those parts are in the country, and we know we need to ship them somewhere else. So the days of doing all those things that waste time are gone.