Re: [LUG] Raspberry Pi Ventilator

[Eion MacDonald <maceion@xxxxxxxxx>]

On 14/04/2020 10:26, Giles Coochey wrote:
> 
> On 14/04/2020 09:51, Michael Everitt wrote:
>> On 14/04/20 09:46, Eion MacDonald wrote:
>>> On 13/04/2020 21:08, Gordon Henderson wrote:
>>>> On Mon, 13 Apr 2020, Roland Tarver via list wrote:
>>>>
>>>>> Hi Folks,
>>>>>
>>>>> BBC News - Coronavirus: Raspberry Pi-powered ventilator to be
>>>>> tested in
>>>>> Colombia
>>>>>
>>>>> https://www.bbc.co.uk/news/technology-52251286
>>>>>
>>>>> Thought this was pretty cool. If it can be made to work.
>>>> Bloody hell, I hope not.
>>>>
>>>> This is the job for a proper embedded microcotroller designed for the
>>>> task. PIC or AVR style not some educational computer running a
>>>> multi-user, multi-tasking operating system vulnerable to 1000 hacks )-:
>>>>
>>>> (And Yes, I have experience of this)
>>>>
>>>> Gordon
>>>>
>>> Can the OS on a R Pi be 'locked' to a single use and hardened?
>>> E.G. If on a machine not connected to internet, is this not just an
>>> 'air-gaped' computer?
>>>
>>> Similar to systems used in nuclear industry in UK , where very 'usual'
>>> PCs are air gapped inside a Faraday cage.
>>> (Disclosure: experience of using such machines.)
>>>
>> The 'best of a bad bunch' would probably be to put an RTOS on the bare
>> metal, and ditch any networking unless really necessary...
>>
>>
> Might be worth pointing out that this is being deployed in Colombia,
> which makes the NHS ventilator situation look like a Exclusive VIP state
> (I know about this as my father had to use private healthcare, but still
> died in this part of the world), and a £25 raspberry Pi is a month's
> wages for most people there. So paying £100,000+ for a Siemens type
> ventilator when you might need 5000 of the things is out of the
> question, if the Pi saves lives it is a good thing.
> 
> To give an indication of the situation, the doctors came to us and said,
> "your father needs a transfusion", we said fine, and had an awkward
> moment, the thing is they didn't have the blood, and wouldn't source it
> themselves, so they lent us a cool-bag with an ice pack and we had to go
> round the city to go and find and buy human blood for the transfusion.
> We did find a couple of pints at the red cross, as it happened, they had
> four matching his blood type, but would only sell us two, despite us
> trying to barter and haggle. $USD50 per pint, in case you are wondering.
> 

"Adequate" for situation is always best.

Example: A flame sterilised army clasp knife to do emergency first aid
and then long rough journey across desert to nearest hospital.
All done under radio instruction from hospital.

Remove skin, patch with bandage, transport to hospital.
 [To take phosphorous impregnated skin off of a  man's back when
phosphorous  grenade wounded the man rather than motor 4 days to nearest
  base hospital to do repair in hospital, by which time he would have
died from phosphorous burn to vital organs(lungs etc)] It worked, but
soldier lost power of speech after being skinned in desert by me. His
"Bloody Back" survived journey and was repaired after reaching hospital.

In poor unequipped countries you do what you can, within resources, so I
applaud the work going into this.
They will do in country trials, of machine with no human connected, then
human connected, so it will be clinically trialled in country, before
live use in wards of sick in say 3 months time.

If living there, I would prefer *something rather than nothing*.
No ability to buy NHS high spec, high cost stuff.


-- 
regards
Eion MacDonald

-- 
The Mailing List for the Devon & Cornwall LUG
https://mailman.dcglug.org.uk/listinfo/list
FAQ: http://www.dcglug.org.uk/listfaq