Document Type
Conference Proceeding
Publication Date
7-2010
Publication Title
The 21st IASTED International Conference on Modelling and Simulation (MS 2010)
Abstract
Medical embedded systems are capable of recording vast data sets for physiological and medical research. Linear modeling techniques are proposed as a means to explore relationships between two or more medical or physiological signal measurements where a causal relationship is believed to be present. Multiple regression is explored for use in medical monitoring, telehealth, and clinical applications.
Spectral regression methods for high-bandwidth medical and physiological signals are demonstrated. The twostage method consists of performing an FFT over a timelagged window of the predictor signal, and constructing a model based on the FFT coefficients. The output of the regression is used in a clustering to explore structure in the array of spectral predictors. It has been applied to medical and physiological time series data, specifically the link between respiration and blood oxygen saturation percentage in sleep apnea patients.
Spectral predictors achieved a dramatically better goodness of fit than time-lagged predictors according to standard analysis of variance measures. In the dataset examined, the spectral model achieved a multiple R2 of 0.90, indicating that 90% of the variation in the dependent signal was captured by the model, while an ordinary distributed lag model had a R2 of only 0.016.
Keywords
Biomedical Modelling, Cardiovascular Modelling, Time Series Analysis, Respiratory Mechanics
Recommended Citation
Macbeth, Jamie and Sarrafzadeh, Majid, "Health Econometrics: Respiration- Oxygenation Correlation through Spectral Models" (2010). Computer Science: Faculty Publications, Smith College, Northampton, MA.
https://scholarworks.smith.edu/csc_facpubs/373
Comments
Archived as published.
Presented at The 21st IASTED International Conference on Modelling and Simulation (MS 2010), Banff, Alberta, Canada, July, 2010