Exploration of Information in Recurrence Plots to Discriminate Human Heart Rate Variability

Hang Ding (2010). Exploration of Information in Recurrence Plots to Discriminate Human Heart Rate Variability PhD Thesis, School of Information Technology & Electrical Engineering, The University of Queensland.

Attached Files (Some files may be inaccessible until you login with your UQ eSpace credentials)
Name Description MIMEType Size Downloads
s40515386_PhD_abstract.pdf thesis abstract application/pdf 7.22KB 4
s40515386_PhD_totalthesishang.pdf Final Thesis application/pdf 3.99MB 13
Author Hang Ding
Thesis Title Exploration of Information in Recurrence Plots to Discriminate Human Heart Rate Variability
School, Centre or Institute School of Information Technology & Electrical Engineering
Institution The University of Queensland
Publication date 2010-11
Thesis type PhD Thesis
Supervisor Stuart Crozier
Total pages 197
Total colour pages 24
Total black and white pages 173
Subjects 08 Information and Computing Sciences
Abstract/Summary Recurrence Plot Analysis (RPA), a nonlinear analysis technique, has been applied to Heart Rate Variability (HRV) studies, at increasingly during the last decade. However, ubiquitous noise and non-stationarity in HRV data greatly complicate the interpretation and evaluation of RP; and importantly, they degrade the RPA analysis performance in applications. In order to deal with the problems and to improve the discriminatory power of the RPA in HRV studies, this work investigated the RP properties under the noisy and non-stationary conditions typically found in HRV studies and, furthermore explored the nonlinear RP information to improve the discrimination of different pathological and healthy cardiovascular conditions. In this research, a complete Holter system with hardware and software was developed to provide the research platform; and six relatively independent studies were conducted. Respectively, they further investigated and explored the RP information from the aspects of 1) Euclidean threshold optimization, 2) delay-embedding effect, 3) local RP structure pattern, 4) variation of HRV RP patterns, 5) multivariate RP pattern analysis, and 6) RP pattern correlation. HRV dynamic discrimination was mainly used to evaluate the analysis performance. Real human HRV data from different healthy and pathological conditions are practically applied in the studies; additionally, a wide range of nonlinear dynamics are synthesized to evaluate the nonlinear discriminatory analysis and to explore the underlying nonlinear characteristics of HRV. The research from a new perspective, explored the complex RP structures and introduced five new RP characterization approaches (Study III to VI) to deal with the ubiquitous non-stationarity and noise properties of HRV data. It showed that complex RP structures arising from underlying dynamic control systems can be valuable for dynamic characterization; and revealed that nonlinear RP pattern and pattern properties are linked to many clinical HRV observations. The real HRV case studies conducted in the work may help explore the underlying nonlinear heart rate regulation mechanism within the autonomic nervous system (ANS) and the improved discriminatory power in this research can directly contribute the patient evaluation and disease stratification in practice.
Keyword Chaos
nonlinear dynamic
recurrence plot
recurrence plot analysis
heart rate variability
Autonomic nervous system
Additional Notes 29, 30, 56, 70, 85, 86, 87, 88, 90, 102, 103, 104, 105, 106, 107, 108, 119, 122, 136, 137, 138, 139, 159, 161

Citation counts: Google Scholar Search Google Scholar
Created: Fri, 05 Nov 2010, 16:27:49 EST by Mr Hang Ding on behalf of Library - Information Access Service