Have a personal or library account? Click to login
Extraction of cardiac and respiration signals in electrical impedance tomography based on independent component analysis Cover

Extraction of cardiac and respiration signals in electrical impedance tomography based on independent component analysis

Journal of Electrical Bioimpedance
Volume 4 (2013): Issue 1 (January 2013)
By: T. Rahman,  M.M Hasan,  A. Farooq and  M. Z. Uddin  
Open Access
|Oct 2013

Figures & Tables

Fig. 1

Basic EIT arrangement.
Basic EIT arrangement.

Fig. 2

Independent Component Analysis block diagram where blind source sˆ (t) is estimated from a mixing matrix A. s(t) are the sources. x(t) are the recordings, sˆ (t) are the estimated sources. Also A is mixing matrix and W is un-mixing matrix.
Independent Component Analysis block diagram where blind source sˆ (t) is estimated from a mixing matrix A. s(t) are the sources. x(t) are the recordings, sˆ (t) are the estimated sources. Also A is mixing matrix and W is un-mixing matrix.

Fig. 3

(a) The original time course of impedance change of a subject during spontaneous breathing with no filtering applied. (b) The Fast Fourier Transform (FFT) power spectrum of this signal showing the frequency characteristics [15].
(a) The original time course of impedance change of a subject during spontaneous breathing with no filtering applied. (b) The Fast Fourier Transform (FFT) power spectrum of this signal showing the frequency characteristics [15].

Fig. 4

Separation of Cardiac and Respiration signal
Separation of Cardiac and Respiration signal

Fig. 5

Proposed method to Separate Cardiac and Respiration signal; After preprocessing ICA is applied on observed signal x to get the respiration template function RT from a set of templates T1 and identically cardiac template function, CT is obtained from remain signal. Unwanted signals are not taken into consideration.
Proposed method to Separate Cardiac and Respiration signal; After preprocessing ICA is applied on observed signal x to get the respiration template function RT from a set of templates T1 and identically cardiac template function, CT is obtained from remain signal. Unwanted signals are not taken into consideration.

Fig 6

Independent Component signals generated from preprocessed raw EIT data.
Independent Component signals generated from preprocessed raw EIT data.

Fig. 7

Selected IC’s (left column) and corresponding reconstructed images (right column)
Selected IC’s (left column) and corresponding reconstructed images (right column)

Fig. 8

Selected IC_9 with its magnified image (a) and its FFT (b).
Selected IC_9 with its magnified image (a) and its FFT (b).

Fig. 9

Selected IC_3 with its magnified image and it’s reconstructed Image
Selected IC_3 with its magnified image and it’s reconstructed Image

Fig. 10

Selected IC’s from observed cardiac signals.
Selected IC’s from observed cardiac signals.
DOI: https://doi.org/10.5617/jeb.553 | Journal eISSN: 1891-5469
Journal RSS Feed
Language: English
Page range: 38 - 44
Submitted on: Feb 1, 2013
Published on: Oct 5, 2013
Published by: University of Oslo
In partnership with: Paradigm Publishing Services
Publication frequency: 1 issue per year
Keywords:
Biomedical Engineering,
Electrical Impedance Tomography,
Independent Component Analysis,
Image Reconstruction,
Cardiac Signal,
Respiration Signal
Related subjects:
Engineering,
Bioengineering and biomedical engineering,
Biomedical electronics,
Physics,
Spectroscopy and metrology,
Life sciences,
Biophysics,
Medicine,
Biomedical engineering

© 2013 T. Rahman, M.M Hasan, A. Farooq, M. Z. Uddin, published by University of Oslo
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.

Previous articleVolume 4 (2013): Issue 1 (January 2013)Next article