The first open-source ventilator tested on human patients

ventilation is the standard course of care for ARDS and most covid crisis cases present as ARDS. it’s great. don’t look it up.

in the first few months, ventilation was used extensively as both a last line of defense and a preventative (maybe we can keep your spo2 above 60% and you will therefore have a better chance to survive).

covid presents very strangely for a respiratory infection, especially in regards to spo2 measurement where patients have greatly diminished capacity to carry oxygen but otherwise seem fine—having conscious conversation, walking. there were a couple of months of basically expert trial and error—take what we know of best practices and apply it in a real world setting with no ability to test effectivity other than whether or not your patient lives or dies. this has been awful for hc worker morale, stacked on top of all the other things (bad ppe availability, long hours for months with no end in sight, hospital administration getting paid while they fire hc workers due to lack of elective surgery, etc).

we are intervening earlier in more proven ways now (such as proning the patient with supplemental o2 rather than immediately ventilating). we know what we are doing better than we did in march (when we knew almost nothing). but it is still extremely difficult to intervene once a patient declines past a certain threshhold. ventilation is still being used as the “keep this patient alive and hope for the best”, but something like 50% of patients with ARDS die within 3 years anyway so you really don’t want to need to be vented.

everyone stay safe. i’m getting the war stories from my sister, who is an ER nurse at Vanderbilt—whose narrative has informed this post since i know fuck all about the application of medicine.

> [Software certification] is something that needs to change, because in the future, we will have more and more sotware in medical devices.

I think this is wrong thinking. Software which drives safety critical systems needs to be perfect, and that is an impossible bar. Whether or not such a target is achievable, it is much MUCH harder if the software is under continuous development. It needs to take enough time to do it properly, and therefore software for critical systems ends up basically being your MVP. It does only what it must, but it does it right. You really don't get a second chance on safety critical systems.

A bit disappointing that your title suggests it's recent news when no work has been done on that repo apart from Readme updates since April. A couple of open-source ventilators have been tested on human patients in France (where this project originates from), all during the same very short period, but none were approved by the health authorities.

I would have appreciated at the very least a description of the project and how it's different from other open-source ventilator projects that can be found on GitHub. Maybe that's included in the link you shared in the comments, but reddit's not pinterest, personally posts that are basically a baitclick title and a link to "more on this on my website" annoy me because I feel they don't enable discussion, they are empty self-promotion.

For those who would like to know more on the rush to develop open-source ventilators due to the current pandemic, this research paper could be a good starting point: https://f1000research.com/articles/9-218 I'm afraid I don't have anything more recent because I've stopped doing a serious watch after May. I have no doubt some of the projects are still moving forward but they have been far less mediatised recently.

No, transportation definitely is not as regulated in the USA as it is in the EU. I'm not against regulation, and you're right that medical devices should be regulated more than less, but tranportation is overregulated for nothing.