Fred has some good ideas for determining the structure of your image
database. One problem with commercial systems is 1) high cost of
customization, and 2)often difficult (or costly) to link to home made
databases such as Access. Our lab system is built in Access and we have not
found a good commercial system for linking to our database without incurring
very high cost (although some of the suggestions from this thread look
promising). We have built an image management system in our database,
complete with thumbnail images, but as Fred said, you need to have the
thumbnails available seperately, which means you have to go through the
trouble of making them. Ideally you would put only your unique case
identifer on the image and then link to the main database for all info about
the image.

(If you want to build a real database I would recommend learning a
realational system like MS-Access first and skip the Excel part. Relational
models are much more powerful and flexible and Excel files often need to be
split up after you transfer them. So in my view it would be better to start
out with the real thing in the first place. Just my 2 cents from long
experience of transferring old Excel files to databases!)

For image storage we have purchased a PowerFile dual-drive CD/DVD jukebox
that holds 200 cd's or DVD's (http://www.dvdchanger.com). This gives us up
to roughly 900 (nine hundred!) GB of image storage capacity. Our unit was
only USD $1,800. An enterprize-level server version is USD $3,600. These can
be daisy-chained (up to 10 units), so you could store up to 2000 discs. They
call this "near-line storage." It is accessable on the network as a shared
drive and is very fast. For images this is fine since they are accessed only
occasionally - not continuously like database info. 

BTW, the PowerFile comes with cataloging software that will show each disk
as a folder and each image as a file. The disc is automatically cataloged
when it is first put in the jukebox. If your discs and images are named, it
will show those names, so for a simple catalog system it may be all you
need. However, there is no search capability beyond the computer system
"FIND" dialog box. 



Tim Morken EMT(MSA), HTL(ASCP)
Centers for Disease Control and Prevention
Infectious Disease Pathology
MS-G32 
1600 Clifton Road
Atlanta, GA 30333

PH 404-639-3964
FAX 404-639-3043
email tim9@cdc.gov 



-----Original Message-----
From: Ian Montgomery [mailto:i.montgomery@btinternet.com]
Sent: Thursday, May 23, 2002 5:56 PM
To: Histonet
Subject: Fw: Image archiving.


Fred,
        Many thanks, I think you've struck the right cord. I already have
the images archived so what I need is an easy retrieval system. CD's can be
labelled into organ systems, animal, technique etc. then details entered
into a suitable Excel system. Will give this a bit of serious thought,
probably reorganize my CD's accordingly then set up a database. The clouds
have cleared and the sun is shining, well maybe a wee bit, this is Scotland
after all and sun is a rare commodity here.
Ian.


Dr. Ian Montgomery,
45 Springfield Park,
Johnstone,
Scotland.
PA5 8JS,
Scotland.
Tel: 01505 335885
e-mail: i.montgomery@btinternet.com
----- Original Message -----
From: "Monson, Frederick C." 
To: "'Ian Montgomery'" 
Sent: Thursday, May 23, 2002 3:32 PM
Subject: RE: Image archiving.


Morning Ian,
Most of the software you are asking about, that is worth any try is
fairly pricey.
If I have it right, you have several to many CD with stored image
files, and what you want to be able to do is find an image, by referencing
one of several to many key words [e.g. Experiment, Investigator, Subject
Tissue, Fixative, Embedment, Coating, Stain, Microscope, etc.].
You mentioned the word "archiving".  I wonder if that is what you
want.  In the days of paper, "archiving" meant that the paper files for a
year were loaded into boxes and sent to a warehouse.  Much work was involved
in recalling such files.  More work the longer the time in storage.
In 'computer talk' archiving often means storage after some form of
compression, as in Microsofts email 'archiving' of "old" messages.  Most of
us do not want our images degraded that way.
I have taken this tack in spite of the fact that you may know all of
the above, because I got the distinct feeling that you already have the
files stored.  What you may need is a method by which to easily access files
WITHOUT having to store them on active hard drives, but rather by
determining where in the growing  library of CD's particular images are
filed.
So, if I haven't completely wasted your time and have, by some
miracle, struck a conversant nerve, I do have a suggestion for a beginning.
You have stored images on CD's without any information, BUT I bet you can go
back to your 'books' and cull all of the relevant image reference data.
If that is true, then my suggestion is this.  Open up a worksheet in
Excel and, using a single image from each image source in your lab, enter
data about that file that you would like to use in searching for it and/or
groups of images, e.g. all thyroids, or all mucous glands in skin of Rana
pipiens sorted by magnification and image source. Each column you establish
should be for some sort of category.  As you can see, for a category such as
magnification the data type will be Integer or Real (if there are mags like
10.5 vice 10,500).
Even if you do NOT wish to establish your own filing system by using
Excel to start, then porting to Access and finally, when it's really large,
porting to MS SQL Server, you will establish criteria for the database you
decide to purchase.  If your wish is to be able to see a small version of
the stored image as a result of a search of such a database, then you will
have to store those images in (or near) the database itself - NOT on CD's.
Everything you want may not be available in boxed software, but it is
imperative that you have some idea of what you NEED!

Hope this helps, and I am looking at the same problem here when we have
digital images arising from SEM, TEM, confocal and other digitalized LM
systems.  I have already started my Excel system even though it is currently
VERY small, I am getting ideas all the time.  For example, what information
in an image header should/can be automatically transferred to a database
entry when the image is saved?

Regards,

Fred

Frederick C. Monson, PhD
Center for Advanced Scientific Imaging
Schmucker II Science Center
West Chester University
South Church Street and Rosedale
West Chester, Pennsylvania, USA, 19383
Phone:  610-738-0437
FAX:  610-738-0437
fmonson@wcupa.edu
CASI URL:  http://darwin.wcupa.edu/casi/
WCUPA URL:  http://www.wcupa.edu/
Visitors URL:  http://www.wcupa.edu/_visitors/


> ----------
> From: Ian Montgomery
> Sent: Wednesday, May 22, 2002 6:47 PM
> To: Histonet
> Subject: Image archiving.
>
>     Can anyone suggest image archiving software. Simple, flexible, user
> friendly and PC based. So far I only have CD's full of labelled images but
> time has now come for some form 'proper' archiving system. All suggestions
> would be welcome.
> Ian.
>
> Dr. Ian Montgomery,
> Histotechnology Unit,
> Academic Support Unit,
> Graham Kerr Building,
> Institute of Biomedical and Life Sciences,
> University of Glasgow,
> Glasgow,
> Scotland,
> G12 8QQ.
> Tel: 0141 339 8855.
> Lab: 6644.
> Office: 4652.
> Fax: 0141 330 5971
> e-mail: ian.montgomery@bio.gla.ac.uk
>
>
>
>
>
>