Back to search

Democritus University of Thrace (DUTH)

Μοντελοποίηση συμπεριφοράς βιοϊατρικών δεδομένων

Abstract

dc:description

This thesis is devoted to the development of Artificial Neural Networks (ANN) and Bayesian Networks (Bayesian Networks), since they can be used in clinical diagnostic areas as suitable classifiers where the mathematical requirements of the parameters that describe the objects are not adequately met. The algorithms and statistical indicators mentioned and implemented in this PhD thesis are pioneers in the field of medical informatics to draw conclusions on various diseases (acute appendicitis) and medical experiments (haemostatic tapes) in order to improve and expedite the treatment of Medical conditions. This thesis attempts the introduction of neural networks RBF and MLP as well as Bayesian classifiers such as Naïve Bayes in hip fracture Bayesian networks are applied to patients with depression in order to find the appropriate model for the fastest diagnosis of the disease. In a set of psychiatric patients, pre-treatment methods are used to enhance the effectiveness of mechanical learning methods. Finally, for the purpose of faster prognosis and prevention of osteoporosis, genetic algorithms, artificial neural networks and pre-treatment methods are used.

Degree

thesis:*
Grantor dc:publisher
Democritus University of Thrace (DUTH)
Year dc:date
2017

Author and committee

dc:creator, dc:contributor.*
Authors dc:creator
  • Galiatsatos, Dimitrios
  • Γαλιατσάτος, Δημήτριος

Subjects

dc:subject × 6

Rights

dc:rights
Statement dc:rights
  • BY_NC
Language dc:language
gre

Identifiers

dc:identifier.*
Identifier
10.12681/eadd/42292

Chain of custody

source
Harvested from
Greek National Archive of PhD Theses
Base URL
phdtheses.ekt.gr/eadd_oai/request
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
citation

Galiatsatos, Dimitrios; Γαλιατσάτος, Δημήτριος. Μοντελοποίηση συμπεριφοράς βιοϊατρικών δεδομένων. Democritus University of Thrace (DUTH), 2017. http://hdl.handle.net/10442/hedi/42292