ePrints.FRI - University of Ljubljana, Faculty of Computer and Information Science

Maja Škrjanc and Mitja Zabukovec (1996) . Prešeren awards for students.

Full text not available from this repository.

Abstract

We have begun to develop new international standardized database of 24-hour ambulatory electrocardiographic recordings which is intended for development and evaluation of nalyzers and algorithms to detect transient ishemic and non-inchomio ST segment changes. For the difference from the existing European Society of Cardiology ST-T database, which contains 2-hour recordings with a number of ischemic ST episodes but innficient number of non-ischemic episodes, is the goal of the now database is to become a reference set of well documented ischemic ST segment episodes, non-ischemic ST episodes connected to sudden shifts of the electrical axis of the heart, episodes of slow drift of the reference ST segment zero level, and mixed episodes. Beside this, the new database will serve as a reference set for exhaust investigations connected to differentiation between physiologic mechanisms generating ischemia and its dinamics. We set the criteria for selecting the recordings. With the aim to identify and to differentiate between ischemic and non-ischemic ST events, we visualizd time series of diaylostic parameters and morphologic features ( ST segment and QRS complex Karhunen-Loeve coeffcients ) of the recordings in compact graphic form. We set standard definitions regarding non-ischemic episodes and mixed episodes, and extended the standard protocol of the reference annotations of the European ST-T database. We developed graphic user interfaces for preprocessing of the recordings as well as interactive annotating of ischemic and non-ischemic ST segmen episodes. Annotating of the ST episodes is performed first globaly on tho episode-by-episode basis by using the trends of morphology features. and after that localy on die basis of original electrocardiographic signals. We annotated 19 recordings, of which ST annotations were verified by a cardiologists. We algo proposed a form of accompanied database documentation.

Item Type: Thesis (Prešeren awards for students)
Keywords:
Number of Pages: 48
Language of Content: Slovenian
Mentor / Comentors:
Name and SurnameIDFunction
prof. dr. Franc Jager235Mentor
Link to COBISS: http://www.cobiss.si/scripts/cobiss?command=search&base=51012&select=(ID=4475732)
Institution: University of Ljubljana
Department: Faculty of Computer and Information Science
Item ID: 3745
Date Deposited: 06 Jan 2017 14:28
Last Modified: 14 Feb 2017 08:12
URI: http://eprints.fri.uni-lj.si/id/eprint/3745

Actions (login required)

View Item