[ExI] good bexarotene article

spike spike66 at att.net
Mon Feb 13 03:53:32 UTC 2012


Cool, the temporary open season is working out well for the topic of
bexarotene: no one is abusing it by wandering off on other unrelated stuff
and it is generating lots of discussion on this timely topic.

Read on please:


>... On Behalf Of BillK
...
>...Snowmed is already in widespread use around the world. For example, the
new UK NHS computerised patient records system has adopted the Snowmed
classification system...Unfortunately the coding system involving disease
and all the sub-categories is too complex for most proles to use. Doctors
have years of training to understand what is going on...BillK

_______________________________________________

BillK, here's the Wiki example of exactly what you are talking about:


SNOMED CT provides a compositional syntax[5] for building new concepts. For
example, there might not be an explicit concept for a burn in the palm of
the hand. Using the compositional syntax it could be described as

 413350009 | finding with explicit context | :
 { 246090004 | associated finding | = ( 284196006 | burn of skin | :
   246112005 | severity | = 24484000 | severe | 
   , 363698007 | finding site | = 321472003 | structure of dermatoglyphic
patterns of palm |)
   , 408729009 | finding context | = 410515003 | known present | 
   , 408731000 | temporal context | = 410512000 | current or specified | 
   , 408732007 | subject relationship context | = 410604004 | subject of
record |  }


My doctor is busy, so I suspect she doesn't use this system.  If we want to
find stuff like some unexplained correlation between Alzheimer's and
bexarotene, we need a way to simplify SNOMED, or create something analogous
to a macro language.  We need to harness the recent and surprisingly
competent speech recognition software to translate the comment "patient
presents with severe burn on the palm of the hand" to:

413350009 | finding with explicit context | :
 { 246090004 | associated finding | = ( 284196006 | burn of skin | :
   246112005 | severity | = 24484000 | severe | 
   , 363698007 | finding site | = 321472003 | structure of dermatoglyphic
patterns of palm |)
   , 408729009 | finding context | = 410515003 | known present | 
   , 408731000 | temporal context | = 410512000 | current or specified | 
   , 408732007 | subject relationship context | = 410604004 | subject of
record |  }

Once we get that far, we have a standardized terminology which can be dumped
into a spreadsheet.  Actually this is one of those ideas which is so
obvious, someone somewhere must already be working on it.  All my best ideas
are that way.  From the description, there is already considerable
intellectual investment in SNOMED in the community, so we would likely be
best working with that system rather than inventing a competitor.  I can
imagine this would really take off if we could get a voice to SNOMED code
macro.

spike





More information about the extropy-chat mailing list