Sprint Day 0: An Open Source ventilator project you can believe in.

[MK14]

@all the critics
The reason why ventilators are so sophisticated and complex is that they are designed for versatility. Forced ventilating a comatose or sedated patient is almost trivial. We have figured this out a long time ago. It gets tricky when you have advanced modes like assisted ventilating. A simple machine that squeezes one of those hand operated paramedic ventilators automatically can free up the professional machines for the more complex cases. Also, production isn't at all keeping up with demand. The reason medical professionals are reaching out to the maker community is that they would rather use a simple machine than having patient drown in their own fluids. Yes, in "peace times" a doctor wouldn't even fart in the direction of a homebrew device, but I am active in the opensourceventilator project and it has a surprising number of medical professionals active there.

And if you think the designs worked on are all shite, join and do it better. 

Thanks, that makes sense.
I much better understand what is going on with those ventilator projects, now.

So, make DIY/Homebrew ventilators, quickly, so some/all will be ready to use, in coming weeks. Even if they are not perfect/brilliant.
Rather than making something better/professional, but it may not be ready, until after the worst/big parts of the virus outbreak (too late!).

tl;dr
Make it quick, or the peak of the "virus battle" (overloading hospitals) will be over.

[helius]

Is there a reason that mechanical ventilation is the first resort for advanced pneumonia and not, say, draining the fluid that is causing them difficulty breathing?

[Blitzschnitzel]

Not a doctor but I think the fluid build-up and difficulty of breathing are both just symptoms of the original lung damage. Whatever is worse will be treated first. Off course at some point fluid also becomes the cause of the breathing problems

[trophosphere]

Is there a reason that mechanical ventilation is the first resort for advanced pneumonia and not, say, draining the fluid that is causing them difficulty breathing?

The following is a simplified response:

In most cases, the accumulation of fluid that builds up in the lung from severe pneumonia is in the tissue rather than in the space between the lung and inner chest wall. Think of it this way, rephrasing your question would be "how would one drain fluid from a wet rag using a needle?" The primary purpose of ventilation is respiratory support rather than to remove fluid from the lung. If a patient is having respiratory distress then it would be pertinent to provide a means of helping them breath before they tire out. It's concerning if a patient is breathing 40 breaths per minute. It's alarming if a patient was breathing that fast and then began to slow down as that would indicate impeding respiratory failure.

[Caliaxy]

Is there a reason that mechanical ventilation is the first resort for advanced pneumonia and not, say, draining the fluid that is causing them difficulty breathing?

https://www.propublica.org/article/a-medical-worker-describes--terrifying-lung-failure-from-covid19-even-in-his-young-patients

[EEVblog]

I  think if you absolutely want to contribute right now, there are myriads of other actions you can currently take that will be way more effective. Just MHO.

Like taking groceries to old people locked down in retirement villages.

[HarryDoPECC]

The propublica link hints at a feature that is going to sink lots of the open source designs.  COVID lung disease manifests partly by wet lungs that are hard to move - so need to push gas in and out, but mostly by gross impairment of diffusion of oxygen from gas to blood.  It's maybe not that hard to lash up some motors and levers to squeeze a bag, but it turns out to be REALLY hard to deliver high partial pressures of oxygen into the lung. 
So a bag-squeezer is easy to design but not very much use.  The useful /essential machine (hi O2) is really hard to design/make.
That's why I think these amateur ventilator projects will deliver little benefit, even if they deliver a product. 
All should be starting with very clear target specs that include high enough fiO2.  If that cannot be achieved then project may be diverting valuable resources.

[Siwastaja]

I  think if you absolutely want to contribute right now, there are myriads of other actions you can currently take that will be way more effective. Just MHO.

Like taking groceries to old people locked down in retirement villages.

... while doing the best you can to avoid exposing them, i.e., paramount hand hygiene, disposable gloves while packaging the groceries, better-than-average DIY face masks (in case you are infected yet symptomless), and, avoiding the temptation of chatting with the elderly, just leaving the stuff on their door.

And, don't forget younger people either. Organizing a mass delivery of basic foodstuff would reduce the number of people visiting stores, making it easier for those who do, maintain safe distance to each other, slowing down the infection chains.

Yes, there are many things you can do.

The ventilator issue is solved by people who do, and who know what to do, and how to do it; not by the eager but clueless management people who need to write on EEVblog forum about their project management plans. Note, the project timeline 2-3 weeks; it has to be a small but efficient team where everyone absolutely knows what they are doing. These teams are people who already know each other, and  emerge automatically within the industry. The design work is all about manufacturability and BOM. 3D printing is not solving the issues, you can't print valves, connectors, hoses, pumps, etc.

[filssavi]

I’m sorry guys, I really get wanting to help, but as engineers or aspiring so you need to get a grip on reality...

The whole most open source ventilator projects are a  very naive waste of time and resources at best, and an extremely disgusting way of getting publicity in a time of crisis at worst

Nobody here, (and in other projects as well) has any clue of the human anatomy, physiology and pathology to even start coming up with requirements that should stop you right there and then, you can’t just google for 5 minutes, copy some stuff from around the web and call it a day.

Before even thinking about a design you should crucially not only be able to tell what the requirements are, but also explain why they were chosen (with simple equations, from first principles, not just hand waiving)

The discussion now should not be about volts and bits but pascals and litres per minute

Then we go to the second major roadblock: production
Let’s suppose some sort of god hands you down the perfect ventilator design what do you do then? You just plonk it down on GitHub and then call it a day, hoping that someone will manufacture them?
No you have to go into production, and here 2 problems arise first money, more specifically whose money pay for all of that  and second you will now have to buy rognificant stocks of parts at very short notice, without any previous relationships with manufacturers (to which I say good luck with that)

Lest but not least if the s***t hits the fan and your makeshift ventilator kills someone who’s getting charged with murder? As No amount of waivers, disclaimers and eulas are going to save you from that

[AndyC_772]

Before even thinking about a design you should crucially not only be able to tell what the requirements are, but also explain why they were chosen (with simple equations, from first principles, not just hand waiving)

The discussion now should not be about volts and bits but pascals and litres per minute

That part is surprisingly readily available:

https://www.gov.uk/government/publications/coronavirus-covid-19-ventilator-supply-specification/rapidly-manufactured-ventilator-system-specification

You're right about the liability aspect, though. I've certainly never designed a device knowing very well that people using it will die. Not 'might' or 'could' if something goes catastrophically wrong, they absolutely, definitely will whether the machine works properly or not.

[beanflying]

It remains the issue that this discussion is now way to late, well before you get to competencies of any contributions toward a cobbled together project. And even if you get it together where do you source the bits in quantity in time without plundering existing manufacturers of known products? The time to look at this was when China first started to blow up.

Applying our relevant skill sets or offering time (if you have it available) to existing providers/manufacturers or like the OP has some skills in the medical field then volunteering will have an immediate and positive effect.

This circular discussion apart from being technically interesting will yield Nil products in the time frame now available -30->60 days 

[filssavi]

Before even thinking about a design you should crucially not only be able to tell what the requirements are, but also explain why they were chosen (with simple equations, from first principles, not just hand waiving)

The discussion now should not be about volts and bits but pascals and litres per minute

That part is surprisingly readily available:

https://www.gov.uk/government/publications/coronavirus-covid-19-ventilator-supply-specification/rapidly-manufactured-ventilator-system-specification

You're right about the liability aspect, though. I've certainly never designed a device knowing very well that people using it will die. Not 'might' or 'could' if something goes catastrophically wrong, they absolutely, definitely will whether the machine works properly or not.

This still qualifies as hand waiving, you are just parroting something you found online and I doubt you really understand it
The point is not the number in itself you (or better the project leaders, the ones who call the shots) have to explain why the number are those and not something else higher or lower

This is important not for the numbers in themselves but to show that you really understand the consequences of your design choices and what implications they have on the person you are ventilating

Also keep in ming that we often talk about peoples dying (I’m guilty of that myself in the previous post) but in case of a small malfunction the result might be worse, ventilation is extremely harsh on the lungs, so it stands to reason that in case of an out of spec ventilator the lungs might be damaged so that the person does not die, but needs respiratory support for the rest of their life  significantly decreasing their quality of life

This might even go against their will (do not resuscitate etc), and and might open you and/or the hospital to extremely large damage claims

[bson]

It's a pointless project in the U.S.  If it takes 45-90 days to get N95 mask manufacturing approved under normal circumstances (unrestricted travel etc), then expect a year for a ventilator design and production under current circumstances (federal employees can't travel to inspect a facility without special dispensation for instance).  There are plenty who are looking to manufacture ventilators, and your efforts would be better placed to lobby the White House to order regulatory agencies to adapt.  Basically, they need to stop the academic wankery (if it's not perfect it's not good enough) and look at the best that can be accomplished given timeframes and resources, non-zero benefit (i.e. "something for nothing"), stacking lots of small odds in their favor, accept risk, and simply realize people are going to die no matter what already.  Congress also needs to indemnify makers, otherwise a single defective unit will put them out of business, which is not conducive to making hundreds of thousands of anything as the defect rate will be non-zero.  The regulatory agencies need to accept it's better for one person to be killed by defective equipment than a thousand to die from a complete lack.  President Trump has shown willingness to accept responsibility and override agencies (he did so to start drug trials for example), and we already know what this will sound like when reported by the click-funded outrage industry (CNN et al).  Such a manufacturing effort would make you a giant target and would probably both ruin you and land you in criminal court (the narrative will be you made shitty equipment to profit and your buddies in the White House helped you out) - although you'd likely be acquitted.  Is it worth it to help people? Most of us would say no, it's just not worth the risk getting too close to any of it.  Stick to lobbying (talk is still free).

[helius]

Is there a reason that mechanical ventilation is the first resort for advanced pneumonia and not, say, draining the fluid that is causing them difficulty breathing?

Think of it this way, rephrasing your question would be "how would one drain fluid from a wet rag using a needle?"

Thanks for the reply. I wasn't actually talking about aspirating the chest wall as in pneumothorax etc, but of non-invasive airway clearance techniques like the ones taught to cystic fibrosis patients.

I mention this because all reports I have heard from doctors facing this pandemic is that mechanical ventilation is particularly hard on the body, takes weeks to recover from, and is often associated with comorbidities in older or weaker patients.

[ehughes]

   There are already safe ventilator designs.      We do not need unproven , hacked together stuff.     Introducing an Arduino controlled shitpile is a horrible idea .

What is needed is to help the existing supply chain who can built qualified units.    As we speak there are already qualified CM's   building real units.       

Right now there is more of a shortage of people who can properly administer the vent.     The are ancedotes of hospitals have to get gynocologists to do it....    The shortage of medical staff is real and they are burning out in critical regions

The time is better spent educating people on how to slow the spread.        There is already well proven tech in the pipeline and lots of people working the problem.

[trophosphere]

Thanks for the reply. I wasn't actually talking about aspirating the chest wall as in pneumothorax etc, but of non-invasive airway clearance techniques like the ones taught to cystic fibrosis patients.

I mention this because all reports I have heard from doctors facing this pandemic is that mechanical ventilation is particularly hard on the body, takes weeks to recover from, and is often associated with comorbidities in older or weaker patients.

In cystic fibrosis, the problem is more so with mucus plugging in the airways whereas in pneumonia the main problem is that the alveoli (the tiny air sacs at the end of the respiratory tree) get filled with fluid (pus/exudate). Sure there is increased mucus production in pneumonia but the main contributor to the decrease in oxygenation is the decrease in area available for effective gas exchange.

The main purpose of using a ventilator in a patient with pneumonia is to (1) provide sufficient respiratory support such that they do not tire out by decreasing the work of breathing with the added bonus of decreasing total metabolic demand, (2) increase effective gas exchange by keeping the alveoli open or opening of the alveoli through something called recruitment, and (3) ensure that their respiratory status is not compromised even if the patient is exhibiting a severe enough altered mental status that they are either unable to maintain their airway or are at high risk of aspiration.

Being placed on a ventilator is hard on the body because often they are on the ventilator for a large number of days such that their respiratory muscles weaken. In addition, skeletal muscle elsewhere in the body weaken as well because they are essentially bed-bound due to often needing to be sedated as (I'm sure everyone agrees) having a tube down the trachea is quite unpleasant. That being said, there are people that are tied to ventilators for many years without the need for sedation because they have had a tracheostomy and a connection is made that way.

Nobody here, (and in other projects as well) has any clue of the human anatomy, physiology and pathology to even start coming up with requirements that should stop you right there and then, you can’t just google for 5 minutes, copy some stuff from around the web and call it a day.

I don't know about that. A few of us on the forum are actually practicing physicians. It's just that we don't see any convincing arguments yet to embark on creating something from scratch when the main problem (as far as I can see) is mainly manufacturing output. This would be the same train of thought as the current need for personal protective equipment and toilet paper(in the US at least). To put it in another way, the demand became so high in such a short amount of time that the current manufacturing setup is unable to keep up. Thus one would believe it would be more valuable and quicker to ramp up production vs creating something new, having to go through regulatory approval, and then having to address the aspect of mass production anyway.

[SilverSolder]

[...] Thus one would believe it would be more valuable and quicker to ramp up production vs creating something new, having to go through regulatory approval, and then having to address the aspect of mass production anyway.

This, in spades, as pretty much anyone with industrial/manufacturing experience would tell you.

[Siwastaja]

It's just that we don't see any convincing arguments yet to embark on creating something from scratch when the main problem (as far as I can see) is mainly manufacturing output.

Yes.

The train of thought for them is not a complete fallacy, though. It has a few realistic, even remotely possible elements to hold the otherwise unrealistic story.

There apparently are some short periods when there is a lack of ventilator equipment in some hardly hit areas, because it takes time to increase ventilator manufacturing. This is true. Now, the story goes like this: some super bright team of science nerds in their garage hack together an iffy-looking, but good-enough thing, build a dozen of them, bring them to a hospital, and save a few dozen lives. (Add some dramatic music here.)

This is like any fantasy, it has remote chances of happening, but the fallacy part is that every "open hardware" idiot is writing themselves in the lead role of that story. What they don't realize at all is that starting to build a bureaucratic community, from scratch, with all the broken telephone play with people they don't know on the interwebz, has already failed before it even started.

They almost always realize they can't do it, or even a meaningful part of it. They don't realize that due to this, they are not needed, and that no one who could do it, is unwilling to participate to their ego projects.

If this fantasy were going to happen, it would be with a few actually bright guys and gals, who already freaking know what to do, to just start doing it, like immediately. No "project planning" needed.

Open source or not, is a completely irrelevant aspect, as are the details whether the design uses Arduino or 3D printing, or not. Anyone who gets fixated to such details are automatically proven morons who can't understand much beyond buzzword and popularity contests. While important to some, they don't get the job done.

[Corporate666]

Entertaining thread

We don't have a ventilator design problem, i.e. we lack well-performing ventilator designs.  We have a manufacturing output problem, i.e. we need more of these things.  Who in their right mind thought the solution to the production problem was to design an entirely new unit?  I know, it's largely about making a simpler design that can be cranked out, but all of the brain power being used on these projects would be better utilized solving the manufacturing problem, including alternate solutions at the component or subassembly level where availability constraints limit how many of an existing/proven design can be manufactured.

There is massive, massive untapped (and currently un/under utilized) manufacturing capacity in this world that we could leverage to make ventilators, but we're going with the best ROI at the moment, using big manufacturers like the car companies to make proven designs.

[tszaboo]

OP, honestly, if you want to help, join an open source project, dont start a new one.

Some people think that engineers are pessimistic, and downers. They bring negative energy, with sentences like " cant be done", "unrealistic expectations". The reality is, any engineer who has some experience, on each project see a thousand way a project can go wrong.

And say, you want to design an open source ventilator. What are you going to do, start ordering all the parts for it? Like realistically, electronics parts have some 12 weeks leadtime. Say, you have the same leadtime for medical valves, pumps and so on.
Are you going to order these parts? There are companies who actually know how to make this, waiting in line to assemble as many units as possible (and make as much profit as possible).

[EEVblog]

What happened to our illustrious leader? 

[Bud]

Perhaps realized that should try to "organize steering and planning committees" and sharpen his leadership skills  elsewhere.

[Siwastaja]

NANDBlog has hit the nail here,

I believe there are really some teams (likely within the existing industry, or something close to it) that are truly solving this problem. They also depend on the usual distributors to get the parts they need; similarly how the healthcare authorities are trying to source the masks wherever they can.

If 1000 different "open source" projects are bootstrapped, and 100 of them go far enough to start buying parts, it's possible this creates temporary shortages or delays the sourcing of the parts for those who actually need them - for those who otherwise are going to get the job done.

Don't play with a serious matter.

[Domagoj T]

https://twitter.com/elonmusk/status/1246279106351816704?s=20


That's quite a piece of engineering.

"Bought, but SpaceX is spooling up to manufacture proportional solenoid valves for Medtronic (confirmed as critical today with chair & CEO)"
So it takes even a huge company that is known for their tight cycles and rapid development some time to start making this. Also, it has to be made, a simple ping pong ball valve won't cut it.

[MK14]

What happened to our illustrious leader? 

In my opinion, it's inconsiderate, somewhat rude and unprofessional. To just disappear like that.
The polite thing to do, would be at least to come on here, and explain what has happened. Even if you are throwing in the towel, and have given up.

I didn't want to put the following comment in, as I didn't want to negatively bias the thread.
But I will now.

I would have thought a person who is going to manage, something like this (apart from a forum, is probably NOT the right place anyway, it should really be a suitable pre-existing company or similar, such as a University, possibly).
The person needs to already have some or all of:
Medical knowledge (ideally Medical Electronics/Devices)
Engineering (especially Mechanical and Electronic)
Software (is a bonus)
Ideally understanding of gases/air flow etc. (pneumatics)
Management skills (ideally in a successful business environment)

The list above, could easily be wrong and/or missing items. Even to make the above list correctly, needs knowledge, skills and experience.