PARALLELIZATION OF ECG TEMPLATE-BASED ABNORMALITY DETECTION

دانشگاه آزاد اسلامی اصفهان(خوراسگان)

منو

رکورد قبلی

رکورد بعدی

" PARALLELIZATION OF ECG TEMPLATE-BASED ABNORMALITY DETECTION "

Document Type	:	Latin Dissertation
Language of Document	:	English
Record Number	:	151121
Doc. No	:	ET22913
Main Entry	:	Sherry L. Kratsas
Title Proper	:	PARALLELIZATION OF ECG TEMPLATE-BASED ABNORMALITY DETECTION
Note	:	This document is digital این مدرک بصورت الکترونیکی می باشد
Abstract	:	Approximately 225,000 adults experience sudden cardiac death each year in theUnited States. In order to develop techniques to predict cardiac arrest, long-term study ofelectrocardiogram (ECG) data needs to be done to detect changes in electrical activity ofdiseased hearts. The goal of such study is to determine a set of electrical patterns thatindicate the deterioration of the heart and possibility of cardiac arrest.In the past, limitations of computing power and storage space restricted theduration of long-term studies to several days. However, with today's technologicaladvancement, data collection can be extended to months or years. The goal of this thesisresearch is to evaluate several alternatives for distributing the analysis of ECG data overmultiple processors. Parallel algorithms utilizing Correlation Waveform Analysis(CWA) were implemented to compare individual heartbeats and form heartbeattemplates. The purpose of the templates is to exhibit the different heartbeat morphologiesencountered in the data. The processing is done on a Linux Beowulf Cluster using thestandardized Message Passing Interface (MPI) libraries. In the thesis, the results of fourdifferent parallel approaches are compared, and their performance is evaluated...,....-...,..tested for theQ1 PC1 bus cardBoth these projects mere sofixare des elopment efforts tonards contributing to dlfferentaspects of Roboucs and lZ1echatronics projects m the Controls and Roboucs Group..
Subject	:	Electericl tess
	:	برق
electronic file name	:	TL46141.pdf
Title and statement of responsibility and	:	PARALLELIZATION OF ECG TEMPLATE-BASED ABNORMALITY DETECTION [Thesis]