Mailing List Archive

Mailing List: techdiver

Message Display

Continuing with the discussion of computer bubble detection/grading, Mark
Lefevre (mark@ms*.co*) wrote:
>
> It (the design of automated detection/grading) would not be very easy to do.
> It would be like designing a computer to read student's essays and decide 
> what letter grade to assign or to taste a compound and say whether it is 
> extremely delicious, delicious, very good or good.
>
For sport divers, it may not be necessary to discern between "extremely
delicious" and "delicious". In other words, the clinical system of grading 
bubbles is not necessarily the best system for this application. Sport divers
are mainly interested in how effective their decompression was. Towards that
end, a simplified bubble monitor could tell them whether or not there were NO
bubbles (very effective), a FEW bubbles (marginally effective), or MANY
bubbles (not effective). Consistent cases of a FEW bubbles would be cause to
modify your diving procedures, gases, tables, times, etc. If MANY bubbles were
detected then it's time to do whatever procedure you have setup to handle DCI,
i.e. O2, IWR, evac, etc.

As to whether or not ANY of this is possible, I suggest that it is. Many
complicated behaviors and procedures have been successfully modeled. For 
example, automated systems now exist that can detect manufacturing defects in
printed circuit boards - work that used to be done by humans.

To summarize my points, if humans ears can detect the "noises" that mark the 
passage of bubbles on a doppler detector, then hardware and software can be
designed to detect them too. The real issue is scoring. For sport divers, a 
complicated clinical system of scoring those noises may not be necessary in 
order to be useful and valuable.

regards,
Greg Dawe

Navigate by Author: [Previous] [Next] [Author Search Index]
Navigate by Subject: [Previous] [Next] [Subject Search Index]

[Send Reply] [Send Message with New Topic]

[Search Selection] [Mailing List Home] [Home]