Representative ways to analyze and survey changes in long-term electrocardiographic recordings

The goal of this research is to explore techniques with which long-term physiologic time-series data can be analyzed, so that relevant changes in physiological signals, particularly the electrocardiogram signal, can be captured, processed, quantified and stored. A new experimental model was developed such that the electrocardiogram can be monitored continuously over thirteen weeks. Cardiotoxicity was progressively induced with doxorubicin in a rabbit model, and electrocardiographic progressions from normal state to diseased state were continuously tracked. Automated methods for analyzing the data were developed to manage and control the extensive electrocardiogram dataset. A significant challenge to this work is the sheer mass of data. This experiment generated 180 megabytes per day per rabbit, totaling around 66 gigabytes for the entire study. Classical ECG parameters significant for the evaluation of heart rate variability were calculated by computer for the entire period of the recordings, and visualized with six different methods.

Name thesis:degree_name

MS

Level thesis:degree_level

Thesis

Discipline thesis:degree_discipline

Lane Department of Computer Science and Electrical Engineering

Year dc:date.available

2000

Author dc:creator

• Crihalmeanu, Simona Gabriela

Contributors dc:contributor

• Stephanie Schuckers.

• Electrical engineering
• Biomedical engineering

Identifier

https://researchrepository.wvu.edu/etd/1069

OAI identifier oai:identifier

oai:researchrepository.wvu.edu:etd-2072