[Sidefx-houdini-list] analysing symmetry

jason lam earthShatteringKaboom at gmail.com
Thu Jul 8 13:12:03 EDT 2010


interested in shape symmetry
so we have volume data, and i have data which maps out anatomica landmarks
eg AC joint to nipple both right and left
so ideally i'd like a measure of how similar the surfaces are to each other
currently what i get as data from my software is an error distance from
-20mm to +20mm and a surface area in mm^2 at that error level
I can also get just the raw data

so my thought would be that if the surfaces are very similar in shape, but
different in size, then the distances might be around 10mm iwth very little
variance
however if the breasts are smilar in volume but one is wobbly in shape, then
the distances would vary a great deal
hence i thought a standard distribution or a 'confidence interval' style
statistic would be good

ie the mean distance was 10mm with a 95% CI of -1 to +22
would suggest very assymetrical breasts



On Fri, Jul 9, 2010 at 2:32 AM, Andrew D Lyons <tstexture at gmail.com> wrote:

> What are they most interested in for their single statistic?
> Difference in volume? Offset in a specific plane?
>
> On 8 July 2010 09:56, jason lam <earthShatteringKaboom at gmail.com> wrote:
> > a single statistic would be great
> > we have the ability to do a colour map of the differences
> > i was playing around with simply doing a standard distribution figuring
> that
> > the smaller it is, the more similar the surface, but a statistician i had
> a
> > chat to said that it wasn't correct
> > He suggested logistic regression analysis, but I'm obviously not
> > particularly strong on statistics...
> >
> > Jason
> >
> > On Fri, Jul 9, 2010 at 2:18 AM, Andrew D Lyons <tstexture at gmail.com>
> wrote:
> >
> >> What's output data are you being asked to provide? One single
> >> divergence statistic? A visualization? A set of key divergences?
> >>
> >> On 8 July 2010 09:42, jason lam <lam0127 at flinders.edu.au> wrote:
> >> > Hi all,
> >> >
> >> > bit random, i've been a bit of a lurker for many years, having moved
> into
> >> > medicine now, but I'm doing some research into breast symmetry post
> >> > reconstruction using 3d laser scanning and could do with a bit of help
> >> >
> >> > Essentially i mirror the breast and align to skeletal landmarks and
> then
> >> > find the distance between the 2 surfaces
> >> > I need some help with the descriptive statistics to be able to
> describe,
> >> in
> >> > a simple numerical way, the similarity of the surfaces and thus the
> >> > symmetry.
> >> > Was wondering if somewhere out there, there was someone who could
> >> possibly
> >> > help?
> >> >
> >> > Best regards
> >> >
> >> > Jason
> >> > _______________________________________________
> >> > Sidefx-houdini-list mailing list
> >> > Sidefx-houdini-list at sidefx.com
> >> > https://lists.sidefx.com:443/mailman/listinfo/sidefx-houdini-list
> >> >
> >>
> >>
> >>
> >> --
> >> =======================================
> >> Andrew D Lyons | Digital Artist | http://www.tstex.com
> >> =======================================
> >> _______________________________________________
> >> Sidefx-houdini-list mailing list
> >> Sidefx-houdini-list at sidefx.com
> >> https://lists.sidefx.com:443/mailman/listinfo/sidefx-houdini-list
> >>
> > _______________________________________________
> > Sidefx-houdini-list mailing list
> > Sidefx-houdini-list at sidefx.com
> > https://lists.sidefx.com:443/mailman/listinfo/sidefx-houdini-list
> >
>
>
>
> --
> =======================================
> Andrew D Lyons | Digital Artist | http://www.tstex.com
> =======================================
> _______________________________________________
> Sidefx-houdini-list mailing list
> Sidefx-houdini-list at sidefx.com
> https://lists.sidefx.com:443/mailman/listinfo/sidefx-houdini-list
>



More information about the Sidefx-houdini-list mailing list