Brown University School of Engineering

Designing BrunO2 for the ventilator shortage

April 9, 2020

A team of engineering faculty, students, alumni and other collaborators are designing and creating prototypes for low-cost ventilators with a device constructed of 3D printed and off-the-shelf components specifically designed for the COVID-19 crisis.

It started with a simple email about a challenge to make a ventilator. Just as Brown University halted its on-campus studies in the wake of the COVID-19 crisis, Engineering Professor Roberto Zenit opened a link to a Code Life Ventilator Challenge, a joint initiative between the Montreal General Hospital Foundation and the Research Institute of the McGill University Health Centre (RI-MUHC). “And I immediately thought of Dan [Harris],” Zenit said, referring to Assistant Professor Dan Harris, “because he has done a lot of work with his rapid prototyping projects in his research and I knew that his lab members are very skilled in that respect. We had a (School of Engineering) faculty meeting that day, and I invited everybody to join, and of course Dan was the one who replied immediately. And that’s how it started.” 

Before the March 31 challenge deadline arrived, three more engineering faculty members, five engineering undergraduates, three graduate students, three physicians, three medical students, two practicing engineers, a research staffer, and a postdoctoral researcher had come together to enter the two-week sprint to collect the best ideas from anywhere in the world to design a life-saving ventilator. The global innovation challenge called for teams to design an efficient-to-manufacture and easy-to-maintain ventilator which could be deployed anywhere needed to save lives.

The BRUN02 TeamThe BRUN02 Team
The team worked tirelessly over spring break, and the following week, to develop its low-cost ventilator design, which they dubbed BrunO
2. The design uses 3D printed mechanical components and open-source electronics for resource-limited settings. It is a pressure-controlled, time-triggered device that is capable of delivering a tidal volume at adjustable respiratory rate and inspiratory ratio, using hospital air and oxygen lines, assumed to be at nominal pressures. The FiO2 can be adjusted. Pressures and flow rates can be measured. All components are 3D printed or off-the-shelf.

Forming the team

Zenit and Harris wasted little time reaching out for help from the moment they signed on. “On the technical side, I reached out to (Assistant Professor) Jacob Rosenstein immediately because I’ve worked with him, and I knew his research,” Harris said.  

As part of the challenge requirements, Harris also knew they would need connections in the Warren Alpert Medical School. He had a contact there he had previously worked with on an orthopedic project, that was willing to launch word of the endeavor to his network of people. “It was through these connections of connections of connections that we ended up with the director of the Miriam Hospital Intensive Care Unit, Gerardo Carino, who joined us.” Adding even more depth to the medical side of the project, Dan Dworkis ’05 also offered to help. Dworkis was Rosenstein’s roommate when the duo were undergrads at Brown, and now serves as an emergency room physician, and assistant professor of clinical emergency medicine at the Keck School of Medicine, University of Southern California. 

“Dr. Dworkis is actively treating COVID patients right now, and using whatever limited free time he has to give us advice. That’s beyond incredible,” Harris said.

Farah Laiwalla, M.D./Ph.D. and current research professor from the Nurmikko Lab, and three Alpert medical students, Anastassia Gorvitovskaia, Roja Garimella, and Nicole Thomasian, rounded out the team on the medical side.

“I knew I wanted the best student designers, and the best mechanical designers, so I then went to the car team,” Harris said. “I know from my own experience when I was on a similar team as an undergrad, that they know what they’re doing in terms of design, doing it quickly, and doing very interdisciplinary work,” Harris said. “Will Haddock, Jacob Morse, John Antolik, and Ian Ho are all the students I immediately started doing the design work with. Ian, John and Jacob are also in my lab right now, so I knew they could contribute.” Eli Silver was a later addition to the group of undergrads on the project.

“Then (Professor) Kenny Breuer jumped in to help advise some students on the fluids aspects. He’s obviously a natural person for this. And (Senior Lecturer and Brown Design Workshop Director) Chris Bull, again a natural fit for the group with the rapid prototyping availability, had volunteered immediately after the faculty meeting.”

Garam Lee: Testing the prototype in the Harris LabGaram Lee: Testing the prototype in the Harris LabWith Rosenstein heading up the electrical effort, and the medical team in place, the group only needed a few more specific skill-set persons to fill in key roles. Rosenstein would bring in postdoctoral researcher Chris Arcadia to help in the final days. And Garam Lee was another of those key pieces added at just the right time. The device was ready to be put together, but in the current situation, all the engineering laboratories had been shut down. Harris and Zenit put together a plan for EHS (Environmental, Health and Safety officers) that would ultimately allow the Harris Lab to be classified as an essential research laboratory, doing work related to the COVID pandemic. “But still undergraduates could not go into the lab,” Zenit explained. “So we needed Garam as a graduate student with access to Dan’s lab, and she immediately agreed to help us.”

Harris said, “I didn’t realize at the time how important each of these people were until we were nearly finished, but looking back, there’s no way we could have done this without them.” 

In the final days prior to submission, Senior Technical Assistant Ben Lyons was fabricating components and graduate student Alberto Bortoni, who had previously worked with Breuer on several bat flight projects, came in to assist. “I was still looking for more graduate help to build,” Harris said. “So I emailed Chris Bull, because he had gotten, under a very specific protocol, some limited access to the BDW for graduate students, staff and faculty only. I asked who was around who could help. He suggested Mark Powell." Powell is a biomedical engineering graduate student who works in Assistant Professor David Borton’s Neuromotion Lab, specializing in neurotechnology to improve human lives. Powell led up the physical prototyping of circuits in the final days.

Dave Durfee, Visiting Scholar in Brown Engineering and founder of Bay Computer Associates, and Andy Leonczyk, electrical engineer at Ximedica, agreed to review the design and serve as practicing engineers on the project.

Ventilator need

With potentially 50 percent of the population on the brink of being infected by COVID-19, most sources agree worldwide health care systems will be strained beyond limits. Currently, there are not enough ventilators to save everyone who needs respiratory support, and doctors are having to make agonizing decisions about whom to save. This challenge posits finding a credible design utilizing the widespread rapid manufacturing tools readily available (such as 3D printers and CNC machines) and low-cost computers (smartphones and Arduinos). 

“Of course, there are many ventilators, and you can normally go buy them, but now there is a shortage,” Zenit said. The vast majority of patients who require ventilator aid can recover, “but since it’s a pandemic, you need 10,000 of these in one city, which is probably not the case anywhere in the world. So we need new ventilators,” said Zenit.

Hospital grade medical ventilators are designed to cover many different therapy strategies, such as controlling volume or pressure, among others. The BrunO2 ventilator was designed as simply as possible, and is pressure controlled. “It does not have all the fancy displays, it only changes a set of parameters that were determined by our physician colleagues,” Zenit said. “Most importantly, one should be able to produce it locally. So once completed, there is a package of information that can be downloaded – in Texas, or in South America, or in West Africa – where users should be able to purchase the materials, print the components, and put it together. Ideally, it should be able to be used safely for humans sick from COVID-19 to treat this specific set of conditions.”

Harris added, “Even money aside, traditional ventilators cannot be made fast enough right now. Supply lines are cut because factories are shut. It’s the same idea as the N95 masks. The problem is that many healthcare workers don’t have them. It comes down to supply and demand.”

Open source science

That the challenge itself stipulated all submissions be in the spirit of open source technology was no issue for the fluids researchers from Brown. “This is part of what Dan and I believe firmly. We’re trying to be good citizens of the world.” Zenit said.

Harris said, “This is how Roberto and I both carry out our research. For me, it’s developing devices and making it open source hardware, so that people can just download the files with the paper, and anyone can build our device inexpensively. Others can use my designs for teaching or to build on, and that’s really been one of the highlights for me over the past few years of my research. This was naturally within that spirit. I believe that to make things progress in science, whether its health or our own fluids research, that we should be collaborating globally. People with less resources don’t have to buy some expensive off-the-shelf equipment. They can print it, and it has been calibrated and tested, and that’s it.”  

When emailed about his role in the project, Dworkis said, “As a front line emergency doctor, I know firsthand how deeply important it is to have the right tools to take care of our patients and our communities — in the middle of this pandemic, that means PPE and ventilators. What the BrunO2 team is trying to accomplish — building a low-cost, open-source vent with the idea of making the plans available to everyone will save lives, here and elsewhere, and hopefully will form a base on which others can grow and iterate. 

“As a Brown alum, I'm not at all surprised that the Brown community stepped up to try to make a difference when the world is suffering. It's what we do. I'm honored to be part of this team,” he added.

Project magnitude

Harris said, “The one thing in hindsight that is really nice about this challenge is that it is being led, and the different stages are being determined, by a group of assembled professionals in medicine. I think there are a lot of efforts out there in rapid prototyping, everyone wants to help, but there are only certain things that can actually be implemented and are viable.” 

“We found this on our own team, but even at the higher level of this challenge, having people in medicine and device designers that do this for a living and actually use these things can say ‘if you do this, it’s going to be useless’ or ‘you don’t need this option because COVID patients respond to this particular type of actuation.’ That’s been completely invaluable,” he said. “Having the medical advisors in particular, has been essential and streamlined this project.”

Zenit added “To develop a product like this that is urgently needed, you cannot compete with companies that do this professionally, but you can contribute to solve a specific set of conditions that you need to tackle and you can only do that by having a physician on the team. I knew nothing about ventilators two weeks ago. But, that is our job. We can read technical documents, understand them, learn and take action. In a place like Brown, this happens naturally. Put good, smart, nice people together and good things happen.”  

Submitting the proposal

When submission time arrived on March 31, the team was still making final adjustments. “How should I describe this? It was a lot of fun. Make sure you put that ‘fun’ in quotations,” Zenit laughed. 

“We all worked extremely hard during spring break to come up with a design, some testing, and a set of documents, letters of support, many things. All of which went from zero to having this impressive proposal, and we submitted with 10 seconds to spare.”

“Literally,” Harris chimed in. 

Zenit said, “Dan’s students were frantically trying to finish the drawings and Garam did the video you may have seen.

"I was trying to document everything and Dan was checking on some final specs. But we did it. As for the final submission, I’m very proud of it. It’s really inspiring to see how all these students and faculty come together and do this amazing work.”

(a) Schematic of complete time-triggered pressure-controlled ventilator design: (b-c) Image of prototype in action during (b) Exhalation cycle and (c) Inhalation cycle. (d) Time trace of voltage readings from pressure sensors during typical breathing cycle, correlating to airway pressure (magenta) and volumetric flow rate (yellow)(a) Schematic of complete time-triggered pressure-controlled ventilator design: (b-c) Image of prototype in action during (b) Exhalation cycle and (c) Inhalation cycle. (d) Time trace of voltage readings from pressure sensors during typical breathing cycle, correlating to airway pressure (magenta) and volumetric flow rate (yellow)
Next steps

The team has already heard back that they have advanced to the top 65 of more than 1,000 original entries for the challenge. Next steps will be forthcoming, but the team plans to continue pursuing the project until a ventilator with the challenge guidelines has been completed.

“Until we have a device that can be printed and used safely by doctors and patients, we will continue doing that,” Zenit said. The team has launched a website (https://sites.brown.edu/bruno2/) for continued updates and giving opportunities, and has already been awarded $20,000 in seed funding from the Brown COVID-19 Research Seed Fund, which was established under the Office of the Vice President for Research. Contributions have also come from the The Warren Alpert Medical School of Brown University, and The Miriam Hospital Educational Fund of the Brown Medicine Division of Pulmonary, Critical Care and Sleep Medicine.

Harris said, “We want to support these students. They are the ones who have made this happen. The faculty tried to organize things, made sure we could meet deadlines, but without the students, there would be nothing built. We would still be sitting in our armchairs reading textbooks.”

Zenit agreed, “The most important thing about our work is the students. Without the students, we are like an orchestra director without players, just waving arms. The music is done by the students.”

Harris said, “The nice thing that has emerged from this is the undergraduates’ role. Garam has been able to help and we’ve had a few graduate students and postdocs, but this original core of the car team group, these are undergrads and I think they showed they can do just as high level of research, given the right tools and given the right motivation and direction, as any graduate student. The students are amazing with design and building via the BDW, and I want them to be properly recognized. 

“Another thing is that it’s been inspiring to hear the students. I’ve never been involved with anything remotely medical in terms of development, and to hear the students already using the jargon and having high-level conversations with some of the local leaders in the hospital is just great. These students are so dedicated. I felt like the team is on equal footing, regardless of whether the team member is a professor or undergrad student or head of the hospital. We are all working together, and it feels very equal. At least to me. That is a really nice aspect about the collaboration.”

-Beth James