From: Dennis Ahr Thanks to everyone who responded to my request for information regarding 3D Image Reconstruction Software for Workstations. I have included all the responses I have received thus far (with minimal editing). Each response is separated with a line. Other interested parties might find something useful here.

Summary follows:

NASA's biotechnology division released a 3D recon. package specifically for EM and confocal work about 3 or 4 months ago. It was for US use only and was apparently available free of charge if you could justify its use (and were a US citizen). Perhaps you could contact them regarding this. It was " advertised" on the WWW NASA pg.
Universal Imaging have a package called "metamorph" that I have'nt looked at (asked for more info. and did'nt get a response) that claims to do 3D recon. as well as photomontage work that can integrate images from a CCD camera. It supposedly has a number of other abilities and is available from Universal Imaging Corp., 502 Brandywine Parkway, West Chester, PA 19380.
There is also a SW package called VoxalView (sp.) from Vital Images that does 3D recon. It is, however, quite expensive () $10K). I've asked around about it and been told it is not so good for some specific applications.
We currently have a package that has been custom designed for our purposes (3D recon. from 35 mm EM negs. with automated alignment to do volume and area measurements of smooth muscle cells and associated nerves). We have'nt gotten as far as publishing with it yet, however when we have we intend to have it available for general distribution and use. According to the guy thats developing the program there are a number of useful bits on the WWW, but I think that's for people who have the time and know how to find them (as I certainly would'nt know where to start).

Shaun Sandow,
Division of Neuroscience,
John Curtin School of Medical Research,
Australian National University,
ACT 0200

Ph. (06) 249 4782
Fax. (06) 249 2687
_________________________________________________________________

I have been associated with a couple labs trying to solve this problem, and we at NRL have also investigated the problem for an upcoming program. In a nutshell, there do exist commercial programs out there, particularly geared towards the medical community. However, they typically have a low resolution and cannot handle the more complex structures often seen in materials science. In every case I know of, this has led to (at least an attempted) development of the actual software. Unfortunately, the research group which appeared to be heading this effort, at least in the area of solid-solid phase transformations (headed by Gary Shiflet of UVa) just lost much of their computer support and no longer has a feasible system (as far as I know). We will be working to develop our own system probably within the next 1/2 year because of this.
For the application you describe, the commercially available software may have sufficient resolution. We have not checked out that software, but Dr. Shiflet did before he started developing their own system. You might want to contact him at gjs@gjssgi.ms.virginia.edu for further information.

Dick Fonda [Naval Research Lab, Washington, DC]
_______________________________________________________________________

VayTek, Inc.
305 West Lowe, Suite 109
P.O. Box 732
Fairfield, Iowa 52556
515-472-2227 FAX: 515-472-8131
Christopher MacLean, Ph.D.
vaytek@ins.infonet.net

Software for PC's (DOS/Windows 3.11, and Windows NT), Unix, and Mac's. Their main software for 3D reconstruction is called VoxBlast, they also have assisant software for aligning 2D sections, and they have deconvolution microscope controlling software as well. All of which is designed to be intergrated with each other and Image Pro Plus. I have tried their demo software (available on line at ftp://128.255.88.195,
User: vbdemo, password: 5UZY2fCn [case sensitive]), and I have had demos done for me but I have yet had the funding to get my own copy of the software. The technical support and sales reps are friendly, knowledgable, and very willing to help.

A second set of software for Unix (and PC's?) is VoxelView, but I do not have the contact info available for them.

Richard E. Edelmann
Electron Microscopy Facility Supervisor
352 Pearson Hall
Miami University, Oxford, OH 45056
Ph: 513-529-5712 Fax: 513-529-4243
E-mail: edelmare@muohio.edu
_______________________________________________________________________

I am using Spyglass, which is now distrubuted by Fortner Research (tel: 800 252-6479 or http://www.langsys.com/langsys.
I am analyzing serial sections obtained by SEM and run the application on a Mac (good enough for this purpose). However, Spyglass is also available for UNIX platforms. Excellent software and easy to use (I never used a manual). I typically have 20 sections, each 640*512 pixels and they are rendered rapidly.This software is only for data reprssentation and not for data anaysis. It would be desirable to be able to extract stereological features from any 3D data set, however, this is not build in.

Hasso Weiland
Senior Scientist
Aluminum Company of America
Alcoa Technical Center
Alcoa Center, PA 15069
412-337-3133
weiland_h@atc.alcoa.com
______________________________________________________________________

Have you seen Voxel View? We used it at Texas A&M to reconstruct confocal sections on a SGI system. I think it would fit your app, but like most workstation software - it's expensive.

James C. Long
Manager/Materials Analysis Lab
Electrosource, Inc.
512-445-6606
jlong@bga.com
_______________________________________________________________________

Try the John Steven Groug at Playfair in Toronto.
Contact Judy Trogadis judy@CAMTWH.ERIC.ON.CA or John, john@camtwh.eric.on.ca
He has written the book on this type of thing which is very different that using confocal input data because TEM sections get distorted and misaligned before they are recorded.
Prof. James B. Pawley, Ph. 608-263-3147
Room 1235, Engineering Research Building, FAX 608-265-5315
1500 Johnson Dr., Madison, WI, 53706
JBPAWLEY@FACSTAFF.WISC.EDU
_____________________________________________________________________

If you contact Joachim Frank, he can tell you about the availability of the SPIDER image processing system. I think it can meet your needs since we use it here to do 3D reconstruction by serial sections, thich-section contouring, and other methods. Both Fourier methods and weighted back-projection are used, and projection onto convex sets is also. Once you have a computer file, the program doesn't care whether the data are from film, TV or SSCCD; you will, of course, have to provide the frame-grabber or storage buffer which converts the video image into a digitized file. You can contact Joachim Frank at joachim@wadsworth.org or look at the web page "http://www.wadsworth.org".
Yours, Bill Tivol
________________________________________________________________________

Your're right-most individuals with workstations have developed their own custom software. Jill Gemmill, at the Neurobiology Research Center at University of Alabama at Birmingham, Birmingham, Al (205-934-7111) developed an excellent 3-D reconstruction package on a Silicon Graphics Workstation. The software was used to perform three dimensional serial reconstruction from images generated by photographs of serial sections in an electron microscope. Granted the interface to capture images directly to the workstation via slow scan camera is not there, but the reconstruction software that Jill developed is super. I cannot find an E-mail address for her, but above is a direct line to her facility. If anything, she might be able to help you out a bit.

Kevin McCarthy, Assistant Professor
Department of Cell Biology
Digital Imaging Microscopy Facility
University of Alabama at Birmingham
Birmingham, Alabama 35294
Phone 205-934-9923/9924, Fax 205-934-7029
________________________________________________________________________

Mayo Clinic puts out a software package that can do almost anything you want it to, however the onsite cost is around $60,000.00.
I have a friend and College at our Beckman Research Institute that is going to teach this software to me. Her name is Janet Sinn-Hanlon, and her email address is:
Janet@delphi.beckman.uiuc.edu

* Janet works only part time, so it may take her a few days to get back to you. Tell her I sent you, she would be glad to discuss the software with you. It could be an individual purchase would be much less expensive.

Lou Ann Miller
Microscopic Imaging Lab
College of Vet. Medicine
University of Illinois
2001 S Lincoln Ave
Urbana,Illinois 61801
217-244-1566
lmiller@ux1.cso.uiuc.edu
________________________________________________________________________

If you accept manual tracing of interesting structures on an overlay-layer of imported tif images, you might use 3D - T.O.P. XW (of IMATEC, Neufahrn, Germany) running on SGI (IRIX 5.3). It has powerful interactive visualization capabilites (realtime rotation and zooming in wireframe mode), good reconstruction and shading caps, but so far rudimentary import caps only. A new tracer/import module for different file format import (and autotrace features) is announced by early 1996.
For information contact:
IMATEK
Elektronische Bildanalysesysteme GmbH
Zugspitzweg 12, D-85375 Neufahrn, Germany
phone: (+49)-8165-65882
fax: (+49)-8165-62750
+++ Dietmar Reiter (dietmar.reiter@uibk.ac.at) ++++++++++++++++ +++ Dept. of Zoology and Limnology, University of Innsbruck ++++ +++ Technikerstrasse 25, A - 6020 Innsbruck, Austria +++++++++ +++ ph: (+43)-512-507-6170 (-6161), fax: ..-507-2930 (-2957) +++
..............................................................................