As of March 18, 2020, the COVID-19 pandemic is growing quickly. The Maker, Humanitarian Enginering, and Free Sofware communities are attempting to answer the call. Public Invention is attempting to help the potential shortfall of ventilators, which is uncertain but could mean hundreds of thousands of needless deaths in a worst-case scenario.
Working with NECSI and EndCoronavirus.org, a self-organized project instigated by Bruce Fenton has arisen. It has become clear, however, that there are and should be many technical approaches.
In order to unify this community, some of us at the #response-ventilators-meta channel at the necsi-edu.slack.com, include Nariman Poushin, have decided any practical response will have to focus on clinical effectiveness and reliability testing, because ventilators are life-critical pieces of machinery. In effect, we need an informal certification process similar in spirit only to FDA approval, which can convince a doctor or medical decision maker to deploy unapproved ventilators as an absolute last resort if there is a shortage of ventilators.
The Free Software and Maker communities know how to form teams; our plan is to form teams for design, testing, training, building, and even fundraising. The basic process we propose is captured in the diagram below.
To deploy a ventilator, extensive testing is absolutely required. This repo is an attempt to build a "test fixture" capable of running a 48-hour test on any ventilator design and collecting data on many important parameters. We hope to create a "gold standard" test that all DIY teams can work to; but this project will proceed in parallel with that.
As part of extensive testings, in combination with BreathOfLife (thanks to Jonathan Orban) and with help from Juan E. Villacres Perez, biomedical engineering student at UT, we have create a Google Sheet downloaded as an OpenDocument Calc and Microsoft Excel spreadsheet of user tests which attempt to match the RMVS standard for Rapidly Manufactured Ventilator Systems. (There are checkboxes in the Google Sheet that don't look nice in the other two formats.) The spreadsheet is licenced under CC0, so you can make a copy of it and use it for testing your own ventilators.
I believe that a test lung is meant to provide a resistance to inflation which models a human lung. An example of a purchasable device is:
Allied Healthcare AHP300 Mechanical Test Lung
5cm H20/L/sec. Resistance Test Lung
Note the units of resistance here: Pressure/volume/sec. Presumable a test that used open air would not accurate model a ventilator.
An example of a certifier, including an explanation of its functionality is here: Alnor-4070 Certifier Flow Analyzer
However, we also need to do multi-iteration reliabillity tests. An initial design is here:
The idea is to make a standalone inline device plugged into the airway. It serves a dual purpose as a monitor/alarm when used on an actual patient, and a test devices for testing prototype ventilators. It also allows for burnin.
Electrical block diagram showing VentMon
