Actions for Heterogeneous Recurrence Analysis of Vectorcardiogram Signals

Email RIS file
Bookmark
Report an Issue

Heterogeneous Recurrence Analysis of Vectorcardiogram Signals

Author
Chen, Ruimin
Published
[University Park, Pennsylvania] : Pennsylvania State University, 2018.
Physical Description
1 electronic document
Additional Creators
Yang, Hui
Access Online

Availability

I Want It
I Want It

Finding items...

Graduate Program
Restrictions on Access
Open Access.
Summary
Recent advancement of sensing technologies provides an unprecedented opportunity for screening of Vectorcardiogram (VCG) signals. This type of signals monitor both spatial and temporal cardiac electrical activities along three orthogonal planes of the body. Therefore, they offer a unique opportunity to study the Myocardial Infarction (MI), one of the leading causes of death throughout the world. Traditionally, recurrence analysis is considered for analyzing VCG signals and characterizing MIs. However, this method relies on the assumption of homogeneity and is limited in the ability to reveal hidden patterns of VCG signals. Noted, heterogeneous recurrences are more related to the variations of recurrence states in terms of state properties and the evolving dynamics of VCG signals. Very little work has been done to study the heterogeneous recurrence behavior of the VCG signals. This paper presents a novel approach of heterogeneous recurrence analysis to study the hidden patterns in VCG signals and thereby characterizing the MIs. First, a nonlinear state space is reconstructed and the state space is then recursively partitioned into a hierarchical structure. Next, extracted quantifiers are effectively characterized the state transitions among segmented sub-regions by fractal representation. Then, the Dirichlet process model as a nonparametric method is utilized to cluster the quantifiers. Finally, a multivariate monitoring method is considered to detect MI behaviors in VCG signals according to the orthonormalized variables in each cluster. The proposed methodology is evaluated and validated with real-world data from human subjects. Experimental results show that the proposed approach not only perfectly captures heterogeneous recurrence patterns of VCG signals through the fractal representation but also successfully monitors and reveals the MIs changes in the dynamics of the signals.
Other Subject(s)
Genre(s)
Dissertation Note
M.S. Pennsylvania State University 2018.
Technical Details
The full text of the dissertation is available as an Adobe Acrobat .pdf file ; Adobe Acrobat Reader required to view the file.
View MARC record | catkey: 23431437