Re: Covid test regime algorithm

Path	csiph.com!eternal-september.org!reader02.eternal-september.org!.POSTED!not-for-mail
From	David Brown <david.brown@hesbynett.no>
Newsgroups	comp.programming
Subject	Re: Covid test regime algorithm
Date	Fri, 8 Jan 2021 16:39:49 +0100
Organization	A noiseless patient Spider
Lines	24
Message-ID	<rt9uc6$1vk$1@dont-email.me> (permalink)
References	<c4b26223-93af-4ac7-a7b0-69056c3f82d4n@googlegroups.com>
Mime-Version	1.0
Content-Type	text/plain; charset=utf-8
Content-Transfer-Encoding	7bit
Injection-Date	Fri, 8 Jan 2021 15:39:50 -0000 (UTC)
Injection-Info	reader02.eternal-september.org; posting-host="07c8c9ba34004d54fa37daa88c8d8fac"; logging-data="2036"; mail-complaints-to="abuse@eternal-september.org"; posting-account="U2FsdGVkX19Jt7JAgvKAMW9bqbvRCGvOV3I3wixgqAE="
User-Agent	Mozilla/5.0 (X11; Linux x86_64; rv:60.0) Gecko/20100101 Thunderbird/60.4.0
Cancel-Lock	sha1:jRW6MLNNEeVKauklxhHIl2PtOLU=
In-Reply-To	<c4b26223-93af-4ac7-a7b0-69056c3f82d4n@googlegroups.com>
Content-Language	en-GB
Xref	csiph.com comp.programming:11730

Show key headers only | View raw

On 08/01/2021 13:54, Malcolm McLean wrote:
> When testing large numbers of people for Covid, and expected cases
> are fairly low, you can reduce the number of test required by pooling
> samples. You then take the pools which test positive and do further
> tests to identify the posiitve patients. Pools which test negative
> are known to have no positive cases.
> 
> So the challenge is to devise an optimal pooling / testing regime.
> 
> Parameters to the model are the number of patients, the expected
> number of cases (actual cases will be Poisson distributed), the cost
> of a swab (one swb needs for each patient for each test), the cost of
> a test (incurred for each pool), the number of tests you can carry
> out in each "stage", and the cost of a "stage". (To explain, if you
> can carry out 500 tests in a stage, and your strategy is to test a
> pool of 1000, then test all individually, you will need three stages
> and 1001 tests. 1 test in the first stage, then you have to wait for
> the results, then a thousand tests, but you don't have capacity for
> that, only for 500, so you have to wait for 2 more stages until you
> get the result).
> 

With those numbers, it would surely then make sense to pool 500 in a
test, not 1000.

Thread

Covid test regime algorithm Malcolm McLean <malcolm.arthur.mclean@gmail.com> - 2021-01-08 04:54 -0800
  Re: Covid test regime algorithm David Brown <david.brown@hesbynett.no> - 2021-01-08 16:39 +0100
  Re: Covid test regime algorithm Richard Heathfield <rjh@cpax.org.uk> - 2021-01-08 16:03 +0000
    Re: Covid test regime algorithm David Brown <david.brown@hesbynett.no> - 2021-01-08 18:41 +0100
      Re: Covid test regime algorithm Richard Heathfield <rjh@cpax.org.uk> - 2021-01-08 19:01 +0000
        Re: Covid test regime algorithm David Brown <david.brown@hesbynett.no> - 2021-01-08 20:14 +0100

csiph-web