About
Algum is a web application developed at Carnegie Mellon University for generating Diamond filters using machine learning on marked up images.
Algum was originally developed by Shiva Kaul and was implemented using JRuby on Rails. It has been completely reimplemented by Jan Harkes using Python and the Django web framework.
News
- October 8th 2013 - Algum now has its own hostname
-
The Algum website has moved again — from cumulonimbus.diamond.cs.cmu.edu to algum.cmusatyalab.org. An automatic redirection has been set up on the old site, but check if you have any bookmarks that link to the old name.
We will soon be migrating Algum to a virtual machine running on faster hardware. Now that Algum has its own hostname, we can do this at will, without having to ask you to update your bookmarks again.
- March 22nd 2011 - Added Image-centric views
-
Use tags to identify a subset of available images, after selecting an image a Deepzoom viewer is used to view the image. From this view it is also possible to navigate to any tasks that have labeled the image.
- November 3rd 2010 - Migrated Algum to new server hardware
-
The Algum website has been moved from kohinoor.diamond.cs.cmu.edu to cumulonimbus.diamond.cs.cmu.edu. An automatic redirection has been set up on the old site, but check if you have any bookmarks that link to the old server.
Aside from the unpronounceable name, the new machine has considerably better specifications, 8 CPU cores, 8Gb of memory, and 1TB of available diskspace, plus several empty drivebays to add extra disks as needed.
- September 6th 2010 - Whole slide image upload and viewing
-
It is now possible to upload a zip archive file containing slide data in any format that is supported by OpenSlide (Trestle, Hamamatsu, Mirax, Aperio and generic TIFF formats).
After upload you can double-click on the slide thumbnail to start an embedded OpenSeadragon viewer to pan and zoom over the full slide.
- July 16th 2010 - New Algum website released
-
Tasks, Labels and Image data have been imported from the old site. However any existing users will need to reset their password before they can log in.
You can reset your password by following this link and entering your email address. You will then be sent an email containing your username and a temporary link through which a new password can be set.