National and Kapodistrian University of Athens
Ανάπτυξη υπολογιστικού μοντέλου προσομοίωσης του αναπνευστικού συστήματος του ανθρώπου
Abstract
dc:descriptionOur main intention, through this dissertation, was to develop a computer simulation model for the human respiratory system. First of all, some basic aspects of modelling, simulation and biomodeling are presented. Furthermore, the current international status in the filed is discussed and the advantages of our effort are outlined. Moreover, the basic aspects of respiratory physiology are presented. So, a computer model to simulate and describe human respiratory system has been developed. The purpose of this system is to provide health professionals with a reliable tool they can use in the decision-making process and educate students. This model was developed considering the previous findings about respiratory system and some equations that were constructed to describe specific functions. The computer model utilized an interactive, graphical and user-friendly interface. For the complete organisation of this model the appropriate software packages were used according to the special demands of this development. This simulation environment is called ReSim (Respiratory Simulation) and consists of four parts and a main screen that leads to them. Each of the first three parts corresponds to a basic part of the respiratory and as a whole they simulate this system. In the last part the outcomes of the mathematical models of the system are compared to real data. The comparison is viewed in diagrams. Another target of our study was to develop a model for the prediction of some spirometric values. For this purpose we based on the artificial neural networks that offer such a potential. The basic characteristics of neural networks theory are mentioned in a special chapter in the dissertation. Normal lung function values are conventionally calculated according to prediction equations that use independent variables such as age, gender, height, and weight. In this study a different method is developed for the prediction of FVC and FEV1 parameters, in order to achieve better correlation of the predicted values to the real ones. Using a sample from the Greek elderly population that was separated into two groups (a training and a testing one), a number of artificial neural networks were trained. Considering that men and women were studied separately and that two parameters (FVC, FEV1) were the target of the study, four cases came up. In each case two neural networks were trained using different transfer functions, number of neurons and number of layers. When passing the inputs of the testing data set to the trained networks it was found that the outputs were well correlated with the corresponding measures of the sample. Furthermore, the match with the sample, for almost all the neural networks developed, was better compared to the matches of older studies that estimate parameter values using prediction equations. This high match allows the use of neural networks for predicting FVC, FEV1 and other spirometric parameters. The conclusions extracted from this study indicate the great potentials of computer simulation for biological human systems; the reliable results of neural modelling encourage us to perform further research in the field of prediction. Any shortcomings of our system could be corrected in the near future; pathological cases could be added; if necessary we could perform a closer debugging. All these could be enhanced in a new version.
Degree
thesis:*- Grantor dc:publisher
- National and Kapodistrian University of Athens
- Year dc:date
- 2002
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Botsis, Taxiarchis
- Μπότσης, Ταξιάρχης
Subjects
dc:subject × 14Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/21191