Aristotle University Of Thessaloniki (AUTH)
Σημασιολογική αναπαράσταση και διαχείριση γνώσης υπολογιστικών προβλημάτων ιατρικής
Abstract
dc:descriptionMedical Informatics is the outcome of prolonged research efforts in both fields of Medicine and Informatics. Lately, tighter bonds have been developed between the two sciences resulting in the deployment of information systems in the field of Medicine. Included in the scientific area of medical Informatics are, among others, the development of theoretical models as well as practical tools for the qualitative and scientific support of medical acts, medical education and organization of health services. Moreover, nowadays special concern is granted to issues like explanation, organization and management of medical knowledge. The result of the abovementioned issues has been the rapid development of medical algorithms with the aid of Informatics which constitutes a large part of contemporary research in the field of medical informatics. In addition, through the Internet (web) numerous medical algorithms along with their implementations are provided free for use. However, despite the proven benefits of implemented medical algorithms in healthcare, the lack of organization and standardization of their information as well as the difficulty of their discovery and utilization lead to compromised use of this system by the medical staff. The purpose of the present thesis was the initial development of semantic descriptions of Medical Computational Problems (MCPs) and the management of resulting knowledge. Defined by the term MCPs are medical problems the solution of which is an algorithm based on mathematical or statistical models and is implemented through Information Technologies (IT). Specifically, in order for this aim to be achieved, the following were developed: • The MCP Ontology for the semantic representation of MCP knowledge which it describes as a triad: For the development of MCP Ontology three interacting ontological schemes that refer to each part of the above triad were created. Furthermore, existing ontological schemas were reused or altered like BibTex Ontology - for the semantic description of bibliographic recourses of MCPs, ConOnto Ontology - for the semantic description of software and material concerning development and execution of implemented applications, and UMLS Ontology - for the semantic description of medical terms in the description of a medical problem. • The knowledge retrieval through ontological questions and the utilization of information retrieval methods inside the MCP Ontology. For the quest of MCPs the method developed was an adaptation of the classical Vector Space Model (VSM) in MCP Ontology, based on which the similarity between MCP semantic descriptions and the user questions is calculated. The MCP vectors are created by the implementation of the weights of the UMLS terms acting as the problem indexing terms in the MCP Ontology. Moreover, another algorithm was developed, expanding the user question and based on the creation of a fuzzy thesaurus of the relevant UMLS terms inside the MCP Ontology. • The dynamic semantic composition of a sequence of algorithms managing a certain medical case. Specifically, using semantic rules, the links between different algorithms are created and used in the construction of a Finite State Machine (FSM) from algorithms. The description of a certain medical case via the MCP Ontology by a user constitutes the language of that case which is recognised by a FSM of algorithms with the final algorithm that manages the case as the initial state and the algorithm of initiation by the user as the final state. • A step-by-step methodology for the development of a modular and expandable Knowledge System (KS) for the MCPs. The system consists of four subsystems: (1) The Semantic MCP Repository through which knowledge in the MCP Ontology is entered and renewed, (2) A Query Engine which retrieves knowledge from the MCP Ontology, (3) The Ontology-Based VSM Space which estimates the relevance of the user question to the MCPs inside the MCP knowledge base, and (4) The MCP-Algorithm Composition for Management of Medical Cases via which it is possible to build an algorithmic sequence from the MCP Ontology, aiding the management of a specific medical case. The thesis is concluded with the experimental results from the utilization of the KS for the MCPs, presenting four indicative examples of use focused on their impact on clinical medical practice. In conclusion, the developed methods of display and management of MCP knowledge along with the further utilization of the proposed KS are expected to enhance the dissemination and use of algorithmic solutions in everyday medical practice. Simultaneously, medical research and high quality medical education are bound to be benefited at a considerable level.
Degree
thesis:*- Grantor dc:publisher
- Aristotle University Of Thessaloniki (AUTH)
- Year dc:date
- 2008
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Bratsas, Charalampos
- Μπράτσας, Χαράλαμπος
Subjects
dc:subject × 26- Σημασιολογική αναπαράσταση γνώσης
- Οντολογίες
- Διαχείριση γνώσης
- Ιατρικοί αλγόριθμοι
- Υπολογιστικά προβλήματα ιατρικής
- Συλλογιστική βασισμένη σε περιπτώσεις
- Ανάκτηση πληροφοριών
- Σημασιολογικοί κανόνες
- Semantic knowledge representation
- Ontologies
- Knowledge management
- Medical algorithms
- Medical computational problems
- Case based reasoning
- Information retrieval
- Semantic rules
- Φυσικές Επιστήμες
- Επιστήμη Ηλεκτρονικών Υπολογιστών και Πληροφορική
- Ιατρική και Επιστήμες Υγείας
- Βασική Ιατρική
- Κλινική Ιατρική
- Natural Sciences
- Computer and Information Sciences
- Medical and Health Sciences
- Basic Medicine
- Clinical Medicine
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/23350