Publications
Text queries can be conducted by Author, Title, or Keyword.
QBISM: Extending a DBMS to support 3D medical images
Source: IEEE Conference on Data Engineering 1994 Feb;10(210):214.
Author: Arya M, Cody W, Faloutsos C, Richardson J, Toga AW.
Abstract:
We describe the design and implementation of QBISM (Query By Interactive, Spatial Multimedia), a system for managing and visualizing 3D medical images in support of structural and functional brain mapping. We propose that such a system be built on top of an extensible DBMS, appropriately extended to handle spatial data types, specifically, scalar fields and arbitrary regions of space within such fields. The challenges are (a) to identify and implement the operations necessary to support medical researchers' queries; (b) to provide acceptable response times, despite the large size of medical images (1-100 Mb per image); and (c) to couple the DBMS with a powerful visualization system. We list the requirements of the application, discuss the logical and physical database design issues, and present timing results from our prototype. The major conclusions are that (a) the use of an extensible DBMS simplifies the introduction of spatial data types and operators; (b) the system response time is dominated by the amount of data retrieved, transmitted, and rendered; and (c) the DBMS' early spatial filtering results in significant performance savings.
Source: IEEE Conference on Data Engineering 1994 Feb;10(210):214.
Author: Arya M, Cody W, Faloutsos C, Richardson J, Toga AW.
Abstract:
We describe the design and implementation of QBISM (Query By Interactive, Spatial Multimedia), a system for managing and visualizing 3D medical images in support of structural and functional brain mapping. We propose that such a system be built on top of an extensible DBMS, appropriately extended to handle spatial data types, specifically, scalar fields and arbitrary regions of space within such fields. The challenges are (a) to identify and implement the operations necessary to support medical researchers' queries; (b) to provide acceptable response times, despite the large size of medical images (1-100 Mb per image); and (c) to couple the DBMS with a powerful visualization system. We list the requirements of the application, discuss the logical and physical database design issues, and present timing results from our prototype. The major conclusions are that (a) the use of an extensible DBMS simplifies the introduction of spatial data types and operators; (b) the system response time is dominated by the amount of data retrieved, transmitted, and rendered; and (c) the DBMS' early spatial filtering results in significant performance savings.