| Title: |
Clinical Data Warehousing |
| By: |
Torben Bach Pedersen |
| Advisor: |
Christian S. Jensen |
| Status: |
Thesis defended April 7, 2000 |
Description
The project will deal with Data Warehousing aspects of data from the
Electronic Patient Record, e.g., modeling and analysis, of large,
heterogeneous amounts of data of clinical nature. The basic idea of the
project is that the area of Clinical Data Warehousing has more complex
demands than the traditional areas in which Data Warehousing has been
applied, thus providing the DW area with new exciting challenges. These
demands will be investigated in cooperation with clinical practitioners. In
the Ph.D. study, the following identified challenges are addressed.
 |
Complex Classification Structures: The classifications encountered in
health care are often very complex. It is difficult yet important to support
issues such as non-strict hierarchies and intelligent handling of change and
time, while maintaining ease-of-use properties such as summarizability.
|
|
Continuous Value Data: Data from clinical measurements and tests must be
analyzed, e.g., using advanced statistics, to answer complex clinical
questions. It is must be considered which analysis operators should be
supported and how pre-calculated data and aggregates can be used
efficiently, as well as how uncertainty in the data is handled.
|
|
Dimensionally Reduced Data: In a Clinical DW, the number of factors
describing the patient, i.e., the number of dimensions in multidimensional
database terms, can be very large. To get an overview of the data, it is
necessary to combine related dimensions into fewer, generalized dimensions.
The Clinical DW should support this, thereby providing the user with exactly
the desired level of detail.
|
|
Advanced Clinical Support: The Clinical DW should try to approach the
world of the clinician by integrating concepts and processes from the
clinical world, for example clinical treatment protocols and medical
research, providing, e.g., (semi)automatic quality management or hypothesis
generation using data mining techniques.
|
Further readings:
|
T. B. Pedersen and C. S. Jensen,
Clinical Data Warehousing - A Survey.
In the Proceedings of the MEDICON '98 conference
[.ps]
|
|
T. B. Pedersen and C. S. Jensen,
Research Issues in Clinical Data Warehousing.
In the Proceedings of the SSDBM '98 conference
[.ps]
|
|
T. B. Pedersen and C. S. Jensen,
Multidimensional Data Modeling for Complex Data.
In the Proceedings of the ICDE '99 conference
[.ps]
|
|
T. B. Pedersen, C. S. Jensen, and C.E. Dyreson,
Supporting Imprecision in Multidimensional Databases Using Granularities.
In the Proceedings of the SSDBM '99 conference
[.ps]
|
|
T. B. Pedersen, C. S. Jensen, and C.E. Dyreson,
Extending Practical Pre-Aggregation in On-Line Analytical Processing.
In the Proceedings of the VLDB '99 conference
[.ps]
|
|
T. B. Pedersen, A. Shoshani, J. Gu, and C. S. Jensen,
Extending OLAP Querying to Object Databases
[.ps]
|
|
J. Gu, T. B. Pedersen, and A. Shoshani,
OLAP++: Powerful and Easy-to-Use Federations of OLAP and Object Databases
[.pdf]
|
|
T. B. Pedersen, C. S. Jensen, and C.E. Dyreson,
The TreeScape System: Reuse of Pre-Computed Aggregates over Irregular OLAP Hierarchies
[.pdf]
|
|
T. B. Pedersen, C. S. Jensen, and C.E. Dyreson,
A Foundation for Capturing and Querying Complex Multidimensional Data
[.ps]
|
|
T. B. Pedersen,
Aspects of Data Modeling and Query Processing for Complex Multidimensional Data.
[.pdf]
|
 |
Copyright © 1998 - 2000. All rights reserved.
|
