A Blog by Jonathan Low


Mar 25, 2020

Could A Bunch of Internet Denizens Produce A Lot Of Ventilators?

Maybe not the top-end $50,000 variety, but repurposed, good enough units for people with less dire symptoms so that the good ones can be used for those with real problems.

And in a crisis like this, it would be foolish not to embrace help from smart, talented volunteers. JL

Jemima Kelly reports in FT Alphaville:

“Helpful Engineering”, set up two weeks ago via a Slack group, brings together thousands (3,400 at last count) of voluntary engineers, technologists, scientists and doctors from across the world. There are 40 sub-groups within the project, each of which is working on a different strategy for helping to deal with the coronavirus crisis in an open-source, collaborative fashion. It seems unlikely that big industry will be able to meet demand in the short to medium term. "I doubt it would be a full ventilator but it might be an adaptor,  hacks like that. Ways to stretch the limits of existing hardware."
In February 2007, Canadian biomedical engineering technologist Clarence Graansma set up a blog to document what he called the “Pandemic Ventilator Project”. The banner at the top described its purpose:
This blog documents our attempt to construct a ventilator design for use in a Flu Pandemic that can be made from readily available materials at the last minute. Hopefully, the situation never arrives where this device will need to be used for a pandemic. But just in case. 
Just in case.
Graansma, who was at that time working as a dialysis technologist at a hospital in Ontario, had recently been called into a meeting with the hospital’s infectious control team. They were concerned that the H5N1 avian flu that was at the time spreading across the globe could turn into a pandemic, and that the hospital’s supply of ventilators – which, like much of North America, were being used at about 90 per cent capacity – would not be nearly sufficient to deal with it. 
Graansma, along with his 17-year-old son and a few others, began to work on building a few prototypes for machines that could be made quickly, cheaply and easily – or, to use the true tech-bro terminology, “hacked” – in the event of a pandemic. He also began getting in contact with government and media to try to raise awareness of this future potential problem, as well as writing posts on his blog. But no-one ended up paying too much attention.
Because the pandemic never came. Or one did, in 2009 – swine flu, or H1N1 – but it wasn’t as severe as people had expected (though it did, it should not be forgotten, kill between 150,000 and 575,000 people globally). So people began to lose interest. Over the following years, the numbers of visitors to Graansma’s blog dwindled from about 1,000 a day to about 10.
Eventually, Graansma’s wife got tired of having a load of old medical junk in the basement, so in December he threw out his old prototypes and equipment.

Time for the techno-solutionists to shine?

Graansma might have thrown out his ventilators (which he reckons, if some of the old rubber parts had been replaced, would have still been functional), but in the past week there has been a resurgence of interest in his blog, which has had as many as 50,000 page views per day.
One of those viewers was a British biochemist and software engineer called Alexander James Phillips. Phillips is one of the founding members of a new project that’s working on ways of dealing with Covid-19, who asked Graanmsa if he wanted to join. The Canadian accepted. 
The project, “Helpful Engineering”, set up two weeks ago via a Slack group, brings together thousands (3,400 at last count) of voluntary engineers, technologists, scientists and doctors from across the world. There are now around 40 sub-groups within the project, each of which is working on a different strategy for helping to deal with the coronavirus crisis in an open-source, collaborative fashion (ie, designs and ideas are shared, with no restrictions on who can use them and develop them).
Here’s a map, from the project’s website, showing where its volunteers come from:

Though the sub-groups are working on several fronts, including on face masks and applications for self-diagnosis, the project’s primary focus is on ventilators: most ambitiously building new ones from scratch, but also fixing old ones, making parts, trying to increase the utility of existing ones and even repurposing machines like CPAP ventilators (which are usually used for sleep apnoea, a condition where breathing stops and starts while the patient sleeps) so that they can be used for Covid-19. 
As regular readers will know, we’re not always the biggest fans of techno-solutionism. Tech bro types who think they can save the world by throwing “innovation” or “disruption” at problems that often require economic, social or cultural solutions rather than technological ones don’t really tend to do it for us, whether they’re working “open-source” or not (much of blockchainland, of which we’re often critical, also operates in this way). 
But necessity is the mother of invention, and the world happens to be in dire need of some real innovation (and, yes, “disruption”). And while we frequently argue that the existing systems we have, while perhaps not perfect, are often more fit for purpose than those that are trying to disrupt them, the healthcare system as it currently stands is just not able to deal with the full impact of Covid-19. So we would be foolish to do anything other than welcome the dozens of projects that are working on this across the internet – pretty much all of which, it should be added, are being done on a voluntary basis.
The shortage of medical equipment in Italy has meant that some older and more vulnerable members of society, with a lower – but still meaningful – chance of survival from Covid-19 than the young and fit, have died. Such heart-wrenching decisions are likely to occur on a regular basis elsewhere when the peak of the virus hits much of the rest of the Western world in the coming days, weeks and months.

What can techno-solutionism actually achieve?

One of the earliest members of the Helpful Engineering project is David Galbraith, an architect, designer and “serial entrepreneur” who co-founded Yelp, spent seven years in Silicon Valley working alongside people like Peter Thiel, and is now a partner at VC firm Anthemis. In other words, a bonafide (if somewhat contrarian) tech bro. 
Galbraith got involved in the group after putting out a tweet asking “Who would like to collaborate on the design of an open-source ventilator?”. His attempt to help, at least at the start, was met with some scepticism: 
A couple of people, when I first posted about the ventilator thing, were saying ‘typical tech bro, think you can solve everything,' which is fair comment. The culture of thinking everything can be disrupted, and everything can be looked at by tech people, as opposed to people who actually know what they doing is obviously a bit obnoxious -- like when Elon Musk tried to create a submarine to get people our of that cave.
But that sort of gung-ho attitude is great sometimes, and it’s part of the way I was thinking when I looked at ventilators. 
Galbraith reckons that it is probably overly ambitious for these kinds of projects to aim to build medical-grade ventilators of the kind that can be used in ICU units for the most severe coronavirus cases and which can cost as much as $50,000 a pop. 
Ventilators are incredibly complex bits of equipment, with a huge number of different parts and functionalities required, as you can see if you pass even a cursory glance at the requirements that the UK government laid out on Friday in its “rapidly manufactured ventilator system specification”. The top-end ones require not only (and most importantly) the ability to supply oxygen and remove carbon dioxide from the body, but also electronic monitoring, a back-up battery in case the mains supply goes down, and the ability to provide so-called “positive end expiratory pressure (PEEP)” that helps hold the lungs open so that the alveoli do not collapse.
But Galbraith does think that the kind of open-source project he is working on could bring about some ideas that stodgier, slower-moving industries might not be thinking about: 
I think it would be hopelessly naive to think we could actually design a piece of medical kit with a bunch of amateurs. But maybe a bunch of amateurs can look at a problem and think of some little bits and pieces that will help. 
I doubt it would be a full ventilator but it might be, say, an adaptor, that allows you to run two pipes off one ventilators -- it might be hacks like that. Ways to stretch the limits of existing hardware. 
He also points out that there are historical precedents for products that have been knocked together quickly so as to meet an urgent need, rather than having been built to last, that have proved surprisingly resilient. The pre-fab Nissen huts used during WW2, for example.
Amateurs also might be able to produce machines that could be used by non-coronavirus patients whose symptoms aren’t so acute. That could free up the high-grade ventilators for those who need them, or could even provide help with breathing for coronavirus patients with less severe respiratory problems. 

A screenshot from GitHub, showing a “low-cost ventilator-ish device”

Do we need to cut the red tape? 

A major issue, though, is red tape. To get something approved for medical use in countries like Britain and the US is a long and arduous process that normally takes several months.
Sam Al-Mukhtar, the CEO of East London-based start-up Geomiq, an online platform that matches design engineers with manufacturers and has a heavy focus on medical devices, says the 22-person company has switched all its attention to trying to fight coronavirus. 
Geomiq has been working with German tech group Bosch and two other companies to come up with a prototype for the outer shell of a ventilator (in other words the hardware; you also need software).
Al-Mukhtar says it would take about a week to perfect the design and at that point they could 3D-print about 1,000 ventilator shells per day using medical-grade materials, but that regulation meant that these would not be approved by hospitals in Britain or the rest of developed world. He told us:
Any medical device, before it’s started to be designed, has to go through a life cycle process with something called a risk register. In that regulation it states that you must start working to this regulation before you start designing and hacking components together. So you can’t release a product to market if you started by hacking it together rather than with the regulation 
My opinion is that the standards are going to have to be dropped. 
Graansma agrees: governments are going to need to temporarily drop their standards if the demand for ventilators is to stand any chance of even coming close to being met:
There are a lot of people putting a lot of time into this, and they’re not doing it for themselves. It could all go to waste if the government does not provide some sort of mechanism that allows these to actually be used. The patients will lose and so will all the people developing this. 
The Helpful Engineering project will eventually come up with a design that’s pretty good, and that will be released. Then people can start making those wherever they live, but whether they can be used or not is down to the government.
He also told us that doctors in the Helpful Engineering group are saying that even if ventilators are not up to medical standards, they still want them. The issue, though, is legal liability:
We have had doctors telling us that if we make them and they’re available and we can show them that they work then they will use them. Because essentially a doctor has a choice of telling somebody that maybe you have a 10 per cent risk of dying from a complication on this ventilator, but if you don’t use it you’re going to die for certain. 
But the issue there is that somebody would have to waive liability issues for that.... We need suspension of liability. 

A motley “bunch of internet denizens”

Another major challenge is organisation across so many disparate groups: Helpful Engineering is by no means the only online project working on such things. The “Open Source COVID 19 Medical Supplies” Facebook group, set up on March 11, has around 38,000 members, for instance. And Geomiq itself has set up an “Engineers Assemble” project that now has almost 300 engineers from several different companies working together. 
Phillips, one of the founding members of Helpful Engineering, told us:
It’s a bit mad because we’re being approached by companies with billion dollar turnovers offering us staff and resources. And I’m wondering: why is there no central WHO-type body doing this? Why is it a bunch of internet denizens doing this? None of us have even met each other before and we don’t have all the processes and procedures and problem-solving in place yet.
While Boris Johnson held a conference call with manufacturers after Britain’s heath secretary Matt Hancock said last weekend that “if you produce a ventilator, then we will buy it,” and that “no number is too high”, none of the engineers or technologists who we spoke to for this story, who are working on open-source projects, were invited to the call. Instead, Johnson spoke with the likes of Rolls Royce, Airbus, and Jaguar, with industry having been asked to deliver 5,000 more ventilators within a month. Big German companies, such as carmaker Volkswagen, which has 125 industrial 3D printers, are also considering deploying their resources into manufacturing medical equipment following a government request. 
“Yes, these are the number one guys in industry, with loads of power and resource, but they’re probably the slowest-moving companies too,” said Al-Mukhtar. 
And while ventilator manufacturers like Ireland-based Medtronic are ramping up their production, it seems unlikely that big industry will by itself be able to meet demand in the short to medium term. (Medtronic, apparently, has also been in discussion with Elon Musk about ventilators, even though Musk on Thursday said his companies would only work on ventilators “if there’s a shortage”. Ahem.)
Tech-phobic as we sometimes might appear to be, we welcome – and indeed applaud – the efforts of these thousands of engineers, technologists and scientists who are coming together to try to find “hacks” for the coronavirus crisis. Just, please, can no one start calling this coronatech.


Post a Comment