National and Kapodistrian University of Athens
Σύγχρονες μεθοδολογίες ανάλυσης ιατρικών δεδομένων: εφαρμογή στην αθηρωμάτωση
Abstract
dc:descriptionThe multifactorial nature of atherosclerosis and the absence of a single biomarker inhibit its accurate prediction. Aim of the present study is to investigate whether the prediction of the outcome of coronary angiography can become feasible by using the most extensive -to date- array of established and novel parameters, harnessed by state-of-the-art machine learning methodologies.To this end, 177 subjects who were planned for coronary angiography enrolled to the study. Demographic, clinical, anthropometric, lifestyle, exercise and dietary data were collected. Ultrasonographic indices of heart function, peripheral artery examination and regional adiposity were also extracted. The concentration of common serum markers as well as more than 60 proteins were identified using Luminex xMAP immunoassay.Following statistical analysis, a series of machine learning classifiers was implemented in R. Numerous biomarkers were selected as being highly informative but none could individually predict accurately the outcome of the coronary angiography. On the other hand many of the combined models exhibited perfect (100%) predictive accuracy.The present study identified numerous probable novel biomarkers of coronary artery disease and has formed robust, highly accurate predictive models of the outcome of coronary angiography.
Degree
thesis:*- Grantor dc:publisher
- National and Kapodistrian University of Athens
- Year dc:date
- 2014
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Vlachos, Ioannis
- Βλάχος, Ιωάννης
Subjects
dc:subject × 16- Στεφανιογραφία
- Στεφανιαία νόσος
- Αναγνώριση προτύπων
- Μηχανικής μάθηση
- Coronary angiography
- Coronary artery disease
- Pattern recognition
- Machine learning
- Φυσικές Επιστήμες
- Επιστήμη Ηλεκτρονικών Υπολογιστών και Πληροφορική
- Ιατρική και Επιστήμες Υγείας
- Κλινική Ιατρική
- Natural Sciences
- Computer and Information Sciences
- Medical and Health Sciences
- Clinical Medicine
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/41670