The KMeD project has completed. The KMeD website now serves as a historical record of the findings and accomplishments of the KMeD team.
A part of the National Science Foundation Scientific Database Initiative Grant IRI 9116849
Last updated on June 12th, 2001

KMeD's query language is KMeD PICQUERY+, a menu-driven, tabular language. Key features include:

  • a modular architecture that permits easy integration of data/image processing and visualization packages,
  • support for TEDM's temporal evolutionary operators such as evolves into, fuses into, and splits into,
  • tight integration with KMeD's visualization and cooperative querying facilities.

Common domain-specific queries can be encapsulated into a custom, user-specified application. The application serves as a shell that accepts the user's specifications, then builds a KMeD PICQUERY+ query based on those instructions. The application also provides access to relevant parts of the underlying data model and knowledge base.

Predicate Tables

The screenshot below illustrates a sample KMeD PICQUERY+ predicate table; click on it to see the menu that pops up for the Relation Operator column of the table.

Processing Tables

The screenshot below illustrates a sample KMeD PICQUERY+ processing table. In particular, this table is used for building histograms from retrieved query results.

Custom Applications

The screenshot below displays KMeD's prototype patient finder application, including a menu that hides or shows windows for the application's data model, knowledge base, and KMeD PICQUERY+ tables. Click on the menu to view a sample knowledge base display.

Below is a screenshot of another custom application, KMeD's prototype PACS study finder. Its accompanying display menu has the application's data model selected; click on it to see what the data model looks like.

Return to the previous KMeD page