Skip to Main Content

LOS ANGELES — In early March, Stanford bioengineer Manu Prakash was attending a conference in southern France and becoming increasingly concerned about the coronavirus outbreak, which was then already sweeping through Europe. “I’d seen what was happening in Italy. Coming back to the U.S., it dawned on me that we were not ready,” he said. Once home, Prakash developed Covid-19 symptoms severe enough that he spent a day in the emergency room. (He was not tested and has since recovered.)

Out of caution, Prakash isolated himself away from his family for 20 days, taking up residence in a room where he stores diving gear. It was a good choice. He took one look at his full-face snorkel mask and immediately thought it might be a good candidate to trick out into an alternative for the face shields and N-95 masks that are in such short supply for health workers.


In a state known for innovation — from creating automatic gene sequencing machines to inventing the internet and iPhone — Prakash is one of many scientists and engineers here who have pivoted their labs to help patients and health care workers amid the pandemic. They’re firing up 3D printers and laser cutters to make parts for much needed medical devices, and using engineering prowess honed by designing spacecraft to repurpose plumbing supplies into cheap, effective ventilators. Here’s a look at three of those projects.

Prakash earned a MacArthur award in 2016 for his work on “frugal science” global health projects that include paper microscopes and centrifuges that cost pennies to make. So working on cheap and efficient solutions to health problems was something for which he was well-equipped. But even he is stunned at the speed at which the snorkel project has developed.

“We went from the spark of an idea to filing [an application] for FDA approval in 14 days,” Prakash said. “We’ve never done anything like this.”


Prakash said his entire lab, even graduate students like Laurel Kroo who were deep into other projects or about to defend their dissertations, have “dropped everything and have been doing this 24/7.”

When Prakash tweeted out the idea of the snorkel mask PPE in March, he immediately heard back from John Pearson, an assistant professor of anesthesiology at the University of Utah who quickly became a collaborator and has run clinical tests of the device, called “Pneumask.” Prakash has made all documentation about the project open and collaborators, from snorkel mask manufacturers, to medical device-maker Boston Scientific, to physicians in France and Chile, have joined the effort.

The basic idea is to pair a snorkel mask, which fits tightly to the face and can be easily disinfected and reused, with a 3D printed custom part that can hold a commercially available filter inside the mask’s ventilation port to trap viral particles. Such filters can be used for multiple days, Prakash said. Part of the beauty of the device, he said, is that it can be used with multiple types of masks and multiple types of filters, even HEPA filters commonly used in vacuum cleaners, so that there is no shortage of necessary supplies to make large numbers of the masks.

Prakash said he is working closely with the Food and Drug Administration for guidance, and has been told by the agency that “improvised PPE” items are acceptable when FDA-cleared masks and respirators are not available. He is planning, through donations and a growing network of industry collaborators, to quickly scale up production and distribution of the masks.

In addition to the snorkel mask PPE, his lab is helping develop a machine that can take waste styrofoam and quickly spin it, like cotton candy, into fibers needed to create N95 masks. He’s also working on a “universal remote” that could be used to adjust ventilator settings so hospital personnel do not need to enter the rooms of infected Covid-19 patients as often, and is part of a team of 60 scientists, engineers, and clinicians who created, an effort dedicated to finding safe and proven ways to decontaminate and reuse N95 masks.

“This is not an academic exercise,” said Prakash, who emerged from quarantine just last week. “This is a war happening on a million fronts and sitting back and just watching the chaos unfold is not an option.”

Collaborators snorkel vent
Clinicians and engineers from across the world who are collaborating with Manu Prakash’s Stanford lab model their masks. Prakash Lab/Stanford

Lonnie Petersen, an assistant professor of aerospace and mechanical engineering at the University of California, San Diego, knew exactly whom to call when she started hearing from medical colleagues overseas about the painful clinical decisions being made for Covid-19 patients because of a lack of ventilators. She turned to James Friend, a professor in the same department whose life’s work is using novel engineering approaches to create and improve medical devices.

“Lonnie called and asked, ‘Can we make ventilators? Can we do something to make a difference?’” said Friend. “I said, ‘My group is supposed to be good at making medical devices and what are we doing this for unless we’re doing it for a purpose like this?’”

The project was challenging from the start because UCSD, like many campuses, was closing and ordering all but critical personnel off campus. Friend put in an emergency request to keep his lab open. It was approved by the administration, provided all lab personnel followed social distancing procedures. “We’ll only have one person in my 1,000-square-foot lab at a time, wearing gloves the whole time, and disinfecting the lab each time,” Friend said. “The last thing we need is the people who can make a difference get ill.”

Ventilators are needed not only because so many people are predicted to be infected with the virus and require help breathing, Petersen said, but also because many patients require ventilation for weeks as they slowly recover. Ventilator manufacturers are struggling to ramp up production because parts are in short supply and a national effort to have auto manufacturers reconfigure their assembly lines to create ventilators will likely not yield significant numbers of new machines until May, weeks after models suggest they may be needed.

“What we are working on are last-ditch ventilators to avoid people having no ventilators at all,” Friend said.

The project started with Petersen, who is also a physician, reaching out to clinical colleagues to ask what was the bare minimum they would need in a ventilator to help patients in respiratory distress because of Covid-19. She then turned to Friend and other engineering colleagues to see if they could create what the clinicians needed.

At their core, ventilators are quite basic machines — pumps that move air into and out of lungs — which may be why it seems everyone from Formula One racing teams and vacuum innovator Sir James Dyson to an army of DIY-ers using tire pumps and old car batteries have offered up ideas for stopgap breathing machines. The engineering challenge, however, is to make sure the machines are reliable and do not induce further damage to lung tissue, especially with a disease as clinically complex as Covid-19, said Petersen.

While some manufacturers, such as Medtronic, have put designs for ventilators online so others could build their own machines, many of those plans require specific parts that are not available. “They rely on supplies from China and those supply chains are broken,” Friend said. “We have to use the simplest motors, the simplest parts, the simplest programming. Those are the safest choices right now.”

The 25-member team focused on the most basic of ventilators, those that use inflatable bags that are hand-pumped to force air into a patient’s lungs. To reduce the staff time involved and the risk of exposing medical personnel to the virus, the team found a way to automate the pumping of the bag of an existing mechanical ventilator, using pieces that could be easily cut with lasers or water jets or 3D-printed.

The ventilators have been successfully tested on campus in a lung simulation lab run by anesthesiologist Preetham Suresh. The team is awaiting FDA approval under relaxed emergency use authorization rules. A Southern California division of Collins Aerospace is ready to start building thousands of the ventilators as early as this week if FDA approval is granted, Friend said.

The team, which went from idea stage to seeking FDA approval in about 10 days, is anxious to get production underway. While Friend estimates standard ventilators will become more freely available nationally in about two months, a University of Washington projection suggests the need for ventilators will peak in California and nationally in mid-April.

“We can’t afford to spend years or even months on this,” Friend said. “We have days.”

UCSD Ventilator
The UCSD ventilator in action. It is attached to an artificial lung that mimics conditions for patients with Covid-19. UCSD

Glen Meyerowitz, a first-year electrical engineering graduate student at UCLA, had been closely tracking the new coronavirus since January because his brother is an infectious disease fellow at Massachusetts General Hospital in Boston.

“He mentioned physicians and clinicians in hard-hit areas were needing to triage care and said he was tremendously worried about physicians having to make those decisions here,” Meyerowitz said. “So I started looking into ventilators and how they work and why they’re so complicated and expensive.”

Meyerowitz, who has an undergraduate degree in physics from Yale, is used to tackling complicated problems. For the five years before he entered graduate school this fall, he had worked for SpaceX, most recently on the Crew Dragon spacecraft being designed to carry humans into orbit.

After speaking with his brother, Meyerowitz jumped into the literature on ventilators and started pinging UCLA clinicians with questions. Like the UCSD team, he quickly realized that Covid-19 patients do not require a ventilator with all of the capabilities of the $30,000 to $50,000 machines many hospital ICUs purchase so they can be used for patients with a variety of diseases or surgeries.

“All that flexibility comes at a price, in terms of cost, design complexity, and manufacturing complexity,” Meyerowitz said. “I thought, what if we focused on something not for 99% of patients, but just for Covid-19 patients, which would be much simpler.”

In less than a week, using a few hundred dollars’ worth of supplies from Home Depot, including an air compressor and basic plumbing tubes and valves, Meyerowitz created a working ventilator. “The build time is dramatically less and doesn’t require skilled labor, so these could be rapidly assembled,” Meyerowitz said.

Meyerowitz is now working with UCLA anesthesiologists and hoping to get the device into clinical use within months. He’s seeking FDA approval for the device, determining how his ventilator could be sanitized between patients, creating detailed industrial engineering specifications for mass production, and readying the device for testing at UCLA’s Simulation Center in preparation for studies on patients in UCLA hospitals.

“I’m a little surprised to say that making the ventilator was the easy part,” said Meyerowitz, who is juggling management of the project with his online coursework now that spring quarter has started. “I’m missing a few Zoom classes,” Meyerowitz admitted. “But the professors understand that right now, this is a priority for me.”

  • I admire their efforts and commitments in helping others in times of need.
    I understand the federal govt has bungled almost everything possible in its response but you can’t use an amateur, not clinically tested device in a literally life-and-death scenario. We won’t get us out of this crisis with DIY or GoFundMe. We need centralized, fast, efficient, scalable government response that uses large private sector manufacturing capability, and when the crisis is over, fundamentally change our system.

    • Andy, I agree with you. Without a centralized government response and coordinated effort, it will be very difficult to get out this crises. I think there will be several new technology and ideas that will come out of this crises but it is not time for DIY.

  • It seems like the ventilators are not a good idea for COVID19 patients. More and more evidence suggest not to use mechanical ventilators as this disease is different than other where mechanical ventilators help.
    For mask, I suggest looking for a different material than the pre-existing melt blown felt fiber to make masks. It is the lack of the material(specially in the US) rather than the capacity of making medical masks.

Comments are closed.