Analysis of Open Source COVID-19 Pandemic Ventilator Projects
— Robert L. Read, Keeshan Patel, and Nariman Poushin
(The web-published spreadsheet is easier to read than the screenshot above.)
During this pandemic, everyone is trying to help. Whether temporarily giving up something you love, like going out with friends, or by sewing masks, or by trying to design a ventilator, most everyone is helping.
In the USA right now there is a high demand for mechanical ventilators. The COVID-19 disease causes some fraction of people, perhaps around 2%, to get Acute Respiratory Distress Syndrome (ARDS). This has been accurately called a “very bad lung thing.”
The overall prognosis of ARDS is poor, with mortality rates of approximately 40%.[21] (https://en.wikipedia.org/wiki/Acute_respiratory_distress_syndrome)
Make magazine recently called open source efforts to make ventilators “Plan C”. “Plan A” is the government taking over factories. Apparently that is not going to happen. “Plan B” is industry increasing production; we hope that happens. But the stakes are high. “Plan C” may be required.
In a public GitHub repo, volunteers at Public Invention and EndCoronaVirus.org have compiled a list of resources and all known open source mechanical ventilator projects. This list is changing every day. Anyone can suggest an improvement or update by entering an issue.
In order to allow volunteers to understand the state of these projects, we have analyzed each of these projects along seven attributes to provide information, transparency, and guidance. These values are purely our own opinions, and are not meant to disparage any project. We have focused not on technical cleverness but on the chance that medical professionals would actually have enough confidence in a particular ventilator to deploy these life-critical devices to patients who need them as a last resort. We have articulated a draft process that we hope is holistic and practical for actually testing and fielding these devices.
Each attribute has five defined levels of readiness. The attributes are Openness, Buildability, Functional Testing, Reliability Testing, COVID-19 Suitability, and Clinician Friendliness.
The web-published spreadsheet is easier to read than the screenshot above.
