Publication
Improving diagnosis processes through multidimensional analysis in medical institutions
| Summary: | For many years, decision support mechanisms have been used on medical applications, providing knowledge and practical expertise on specific areas, as well ensuring a better support to medical staff do enhance their jobs. Nowadays, we haven't the need to prove once more their utility in clinical scenarios. Existing applications can easily show that. However, we can discuss some alternatives to improve some of the existing decision-support systems, especially at the data repositories level. We believe that a well designed and implemented data repository could make the difference in the effectiveness and in the quality of service provided by any decision support systems, and especially by an associated data warehousing system. A well organised multidimensional schema, containing every possible dimension of analysis and the necessary evaluation metrics, combined with an effective populating strategy, integrating specific domain oriented extraction, transformation and integration mechanisms, are basic ingredients to dispose a successful data warehouse for a conventional data warehousing system. In this paper, we discuss a possible schema for a data warehouse especially oriented to support medical diagnosis processes, presenting all its basic structures, including multidimensional schemas, fact-tables organization, dimensions of analysis, and some exploitation mechanisms. Moreover, we intend to demonstrate that it's possible to have more effective diagnosis processes if we take into consideration some multidimensional data design and populating aspects in the data structures, that receives diagnosis information, and use adequately an On-Line Analytical Processing system to explore them. |
|---|---|
| Main Authors: | Belo, Orlando |
| Subject: | Medical Decision Support Systems Diagnosis and Treatments Data Warehousing Systems On-Line Analytical Processing Multi-dimensional Data Structures |
| Year: | 2011 |
| Country: | Portugal |
| Document type: | book part |
| Access type: | restricted access |
| Associated institution: | Universidade do Minho |
| Language: | English |
| Origin: | RepositóriUM - Universidade do Minho |
| Summary: | For many years, decision support mechanisms have been used on medical applications, providing knowledge and practical expertise on specific areas, as well ensuring a better support to medical staff do enhance their jobs. Nowadays, we haven't the need to prove once more their utility in clinical scenarios. Existing applications can easily show that. However, we can discuss some alternatives to improve some of the existing decision-support systems, especially at the data repositories level. We believe that a well designed and implemented data repository could make the difference in the effectiveness and in the quality of service provided by any decision support systems, and especially by an associated data warehousing system. A well organised multidimensional schema, containing every possible dimension of analysis and the necessary evaluation metrics, combined with an effective populating strategy, integrating specific domain oriented extraction, transformation and integration mechanisms, are basic ingredients to dispose a successful data warehouse for a conventional data warehousing system. In this paper, we discuss a possible schema for a data warehouse especially oriented to support medical diagnosis processes, presenting all its basic structures, including multidimensional schemas, fact-tables organization, dimensions of analysis, and some exploitation mechanisms. Moreover, we intend to demonstrate that it's possible to have more effective diagnosis processes if we take into consideration some multidimensional data design and populating aspects in the data structures, that receives diagnosis information, and use adequately an On-Line Analytical Processing system to explore them. |
|---|