Biomedical Time Series Analysis
Library Home || Full Table of Contents || Library Help
|Biomedical time series are sequences of observations recorded over time (a simple example would be hourly measurements of the blood pressure of a patient). These series of measurements exhibit characteristics of the systems and mechanisms that generate and control the recorded signal... The greater the correlation is amongst the variables, the lower will be the fractal dimension. While it is correct to think of non-linear dynamical systems in the chaotic mode as having fractal characteristics, these are usually secondary properties (analogous to the distances between the rings of Saturn or the distances between points in a Poincare section through the attractor). Instead of looking for higher dimensional structures to be formed out of a time series, for most biological time series we look for evidence of long memory behavior, expressed as correlation between members of the series that extends over long times. Unified framework: schematic and hypertext graphs; case studies where the analysis methodology is applied to various sets of data.|
|Math Topics:||Fractals, Data Analysis, Medicine/Health, Animal Biology|
© 1994- The Math Forum at NCTM. All rights reserved.