Jeff Potter (jepotter@caregroup.harvard.edu)
Thu, 14 Oct 1999 16:38:04 -0400
----------
>From: bond@stereosgi.neur.cwru.edu (Angus Bond)
>To: Mark Davey <mkd@medphys.ucl.ac.uk>
>Cc: info-volumizer <info-volumizer@corp.sgi.com>
>Subject: Re: Fly-through suggestions
>Date: Thu, Oct 14, 1999, 1:22 PM
>
>> Message-ID: <38061800.A98BE7D5@medphys.ucl.ac.uk>
>> Date: Thu, 14 Oct 1999 10:50:56 -0700
>> From: Mark Davey <mkd@medphys.ucl.ac.uk>
>> To: info-volumizer <info-volumizer@corp.sgi.com>
>> Subject: Fly-through suggestions
>>
>> Hi..
>> I am looking to use Volumizer to effectively 'fly-through' a CT or MRI
>> volume. For instance, this could be used as a teaching aid for
>> endoscopic surgery. Does anyone have any ideas concerning steps that
>> should be taken to do this using Volumizer. My initial attempts have
>> been less than satisfactory resulting in quite blurred, very aliased
>> images.
>> Regards
>> Mark
>>
>> --
>> From: Mark Davey
>> Dept. of Medical Physics, UCL.
>>
>> TEL: +44 (0)171 915 1673.
>> FAX: +44 (0)171 837 9279.
>>
>> Institute of Laryngology and Otology,
>> 330 Gray's Inn Road,
>> London.
>> WC1X 8GE.
>
> Hello Mark,
>
> I do not have the answer you seek, but I can tell you that, if you
> are emulating an endoscope, you can expect to see blurred and/or
> aliased images.
>
> The problem has to do with image resolution. CT and MR scans are
> usually 512 pixels across a field-of-view of 240 mm and the slices
> are usually 1 to 2 mm apart. That means that each voxel is about
> 0.5x0.5x1.0 mm at best. An endoscope usually has a field-of-view
> of about 10 to 20 mm, so you will be looking at a field of about
> 40x40 voxels at best and 20x10 voxels at worst. I can guarantee
> that the result will be blurred and/or aliased.
>
> A good rendering system will produce an anti-aliased image so that
> you get an understanding of what you are seeing. You may also be
> able to get a crude surface rendering. The rendering system cannot
> create data that was not in the original CT or MR scan, so the very
> best it can do is approximate reality. Once you understand this,
> you can find a rendering system that will give you a good
> representation of the volume. For that, you will have to talk to
> the other members of this mailing list.
>
> Good luck!
> Angus
> -----------------------------------
> Angus Bond
> bond@stereosgi.neur.cwru.edu
> bond@modex.com
> +1 216 844-8963 lab/office
> +1 440 834-1414 answering machine/fax
This archive was generated by hypermail 2.0b2 on Thu Oct 14 1999 - 15:46:21 PDT