Re: Modeling the Aural canal and 3D scanning

From: Ernst Mueller (E.Mueller@gom.com)
Date: Fri Mar 08 2002 - 21:43:43 EET


I am not so sure if digitizing is far behind. If you look at:
http://www.gom.com/En/Applications/Digitizing/RapidPrototyping_Milling/magen.html
you see some nice scans, with fine details, with direct interface of the
scan data into RP.

For white light scanning, we need a projection and viewing angle, but
typically what you can see with two eyes, we can scan.

We (and some other digitizer suppliers) digitize silicon castings for
different hearing aid suppliers. In this application it is a cost and
time issue and only with very few samples we can not get a full
coverage.

To see what scanning can do some jobs which are too big for standard RP,
please look at the digitizing of the actual flight test model of the
NASA X38 Recovery Vehicle as given in:
http://www.gom.com/En/Applications/Digitizing/Qualitycontrol/x38.html.

With best regards

Ernst
_______________________________________________________________________
Ernst Mueller mailto:E.Mueller@gom.com
GOM International AG http://www.gom.com
Bremgarterstr. 89B Tel.: +41 (0)56 631 0404
CH-8967 Widen, Switzerland Fax: +41 (0)56 631 0407
_______________________________________________________________________

> Bill Richards wrote:
>
> I must agree with Dr. Bibb: Never stick anything in your ear (except
> your elbow).
>
> Hearing specialists routinely make castings of a patient's ear canal
> with a soft material in order to create a custom fit hearing aid.
> Please go to your doctor to find out where you can have this done
> safely. I can't imagine that anyone would find having their eardrum
> ripped out of their body to be a pleasant experience...
>
> 3D scanning technology is not up to the level that many people would
> like to believe. I would put scanning technology at least several
> years behind rapid prototyping.
>
> Laser scanners can only scan what can be seen on a straight line. Over
> hangs and folds will never be scanned. Metals, transparent materials,
> and materials with translucence or specularity will cause problems
> with the scanner determining just where the light beam actually
> touched the surface.
>
> Mechanical arm scanners can overcome that issue, so long as the probe
> can be worked into the tight places. But the quality of your data
> depends on the number of times you touch your probe to the target
> surface.
>
> Another type of scanner actually destroys the subject, by shaving it
> away layer by layer and imaging each layer. This technique guarantees
> that hidden geometries will be scanned. The only other way would be to
> use MRI or CAT scanning.
>
> For creating a model of an ear, I don't believe you are going to get a
> true scan without using MRI or creating some custom probe-scanner.
> Perhaps an endoscope with a laser scanning from its tip. I think the
> best way is as Dr. Bibb suggested: Have a doctor create a casting of
> your subject's ear, and then scan the casting.
> __________________
>
> Bill Richards
> Delta Search Labs
> T: 617-551-4615
> F: 617-551-4651

For more information about the rp-ml, see http://rapid.lpt.fi/rp-ml/



This archive was generated by hypermail 2.1.4 : Tue Jan 21 2003 - 20:13:34 EET