University of Ioannina
Εξόρυξη πληροφορίας και ιατρικά συστήματα υποστήριξης απόφασης
Abstract
dc:descriptionThe objective of the PhD thesis is the design and development of decision support systems, with knowledge derived using data mining techniques. Specifically, in the field of biomedicine we designed and developed a decision support system for ischeamia, by processing and analyzing the ECG signal. From the biology domain, we designed and developed a method for the classification of proteins into folds. Finally we created a novel methodology than can be applied in several sequential domains, for the creation of sequential classification models, if the data are described as discrete time series. In the first chapter we describe the fundamental issues of the healthcare decision support systems, data mining as well as their interconnection and relationship. A general data mining framework for creating decision support systems is also presented. At the end of the first chapter we provide a detailed literature review concerning the two applications we have developed: ischaemic beat detection and protein fold and class recognition. In the second chapter we present a literature review of decision support systems that have been developed and are widely used in the healthcare domain and provide algorithms that are used for data mining. Moreover, detailed state of the art in the domains of ischaemic beat detection and protein fold and class prediction are included. In chapters 3 and 4 we describe in detail the methods we followed for solving the real world applications – ischaemic beat detection and protein fold recognition and class prediction. The methods follow a general scheme that corresponds to four steps: a) data collection b) data preprocessing c) data analysis using data mining techniques and d) creation of the classification model – decision support system. Special attention is given in the third step, where we mainly used association rule and sequential pattern mining techniques. The results we have achieved show the effectiveness of these methods. In chapter 5 we present a methodology that can be used for the creation of sequence classification models. The methodology integrates sequential pattern mining techniques and optimization algorithms and is compared with two other similar methodologies. Finally in chapter 6 we provide conclusions, remarks and future development of out thesis. A future development that we have already identified is the creation of decision support systems using data mining techniques applied in data clinical and genetic. The integration of these two types of data can help in better describing the situation of the patient and thus in a more accurate diagnosis. The use and integration of clinical and genetic data can help in diseases like cancer. At the end of this chapter we present a system’s architecture that can achieve this integration and create the corresponding healthcare decision support systems
Degree
thesis:*- Grantor dc:publisher
- University of Ioannina
- Year dc:date
- 2009
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Έξαρχος, Θεμιστοκλής
- Exarchos, Themistoklis
Subjects
dc:subject × 16Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/17297