Sridhar P. Reddy

New Signal Processing Techniques for Improved Precision of Noninvasive Impedance Cardiography

Impedance cardiographic determination of clinically important cardiac parameters such as systolic time intervals, stroke volume, and related cardiovascular parameters has not yet found adequate application in clinical practice, since its theoretical basis remains controversial, and the precision of beat-to-beat parameter estimation has until recently suffered under severe shortcomings of available signal processing techniques. High levels of noise and motion artifacts deteriorate signal quality and result in poor event detection. To improve the precision of impedance cardiography, new techniques for event detection and parameter estimation have been developed. Specifically, matched filtering and various signal segmentation and decomposition techniques have been tested on impedance signals with various levels of artificially superimposed noise and on actual recordings from subjects in a laboratory study of cardiovascular response to a cognitive challenge. Substantial improvement in the precision of impedance cardiography was obtained using the newly developed signal processing techniques. In addition, some preliminary evidence from comparisons of the impedance cardiogram with invasive aortic electromagnetic flow measurement in anesthetized rabbits is presented to address questions relating to the origin of the impedance signal.

Coherent ensemble averaging techniques for impedance cardiography

Ensemble averaging provides some improvement in the stability of the signal and signal-to-noise noise ratio under conditions of nonperiodic influences of respiration and motion. Coherent averaging techniques were developed to determine whether further enhancement of the impedance cardiogram could be obtained. Physiological signals were obtained from 16 male and female subjects during resting conditions, while delivering a speech, and while undergoing submaximal bicycle exercise. Results indicated that improved resolution of dZ/dt signal events could be obtained using coherent ensemble averaging. Although some improvement in precision of event location was obtained, most enhancement of the impedance cardiogram occurred in measurement of the amplitude of the dZ/dt maximum (ejection velocity) during speaking and exercise conditions. Validated increases in dZ/dt maximum exceeding 20% were obtained in some subjects with coherent averaging, suggesting that the diagnostic utility of impedance cardiography can be improved using this technique.<<ETX>>

Signal fidelity requirements for deriving impedance cardiographic measures of cardiac function over a broad heart rate range.

Our findings indicate that the impedance cardiogram spectrum extends from DC to 50 Hz. Any amplifier with an upper band limit less than 50 Hz can be expected to produce attenuation and distortion of the impedance cardiogram. This signal attenuation may be systematically enhanced under conditions of high heart rate when a greater proportion of signal energy will be in the upper frequency range of the impedance cardiogram spectrum. Therefore, the present study was designed to assess the influence of amplifier bandwidth on dZ/dtmax, stroke volume, and systolic time intervals (LVET, PEP, QZ, QX). Simultaneously measured delta Z and dZ/dt signals from two impedance cardiographs, with corner frequencies of 120 and 60 Hz for the delta Z and 50 and 15 Hz for dZ/dt channels, were contrasted over a broad range of heart rate (70-150 bpm). In addition to the analog dZ/dt signals obtained from the instruments, the delta Z signals were digitally converted to dZ/dt by off-line digital differentiation with a 50 Hz corner frequency. The results demonstrated that the measurements with the 15 Hz corner frequency, when compared with the 50 Hz corner frequency measurements, systematically attenuated the dZ/dtmax amplitude and stroke volume measurements as heart rate increased. The attenuation of dZ/dtmax and stroke volume ranged from about 13% to 26% as heart rate increased from 70 to 150 bpm. When the upper bandlimit was 50 Hz, the dZ/dt signal had greater resolution of waveform events and produced less prolonged systolic time intervals. The 15 Hz amplifier differentially influenced the B point, Z-peak and X minimum, having no apparent effect on the temporal location of the B point, but delaying the Z-peak about 21.7 ms and the X minimum about 7.4 ms. These findings indicate that impedance cardiographs with insufficient upper bandlimits will differentially influence ICG-derived measurements as heart rate varies.

Improved reliability of impedance cardiography by new signal processing techniques

The authors used the impedance cardiogram (ICG), the phonocardiogram (PCG), and the electrocardiogram (ECG) to determine stroke volume and systolic time intervals. Comparisons between a fully automated PC/AT-based signal processing system and a computer-assisted visual detection processing system revealed comparable means, standard errors, correlations (Spearman r), and reliability (Chronbachs alpha ) coefficients on all relevant parameters. Other analyses confirmed the utility of using Q onset to mark the beginning of ventricular systole, the B point of the ICG to denote the aortic opening, and end diastole as the reference for determining the amplitude of dZ/dt/sub mzx/.<<ETX>>

Class 3 adaptive filters using time domain smoothing

A design of Class 3 adaptive filters using time domain smoothing is presented. These filters require no apriori knowledge about the signal and noise that are being processed. Class 3 filter performance depends on the smoothing used and the signals and noises presented to the transfer functions. Different time domain smoothing windows are considered. Simulations using EEG signals confirm that these Class 3 transfer functions degenerate to classical band-limited filters.<<ETX>>

New signal processing techniques for improved reliability of impedance cardiography

Different techniques have been developed for parameter extraction and segmentation of the impedance cardiogram for improved reliability and precision of beat-to-beat determination of systolic time intervals, stroke volume, and related cardiac indices. The solutions presented improve on previous techniques by: (1) substantially reducing jitter in the localization of events within the cardiac cycle; (2) providing exact determination of signal amplitudes even in the presence of artifacts and interference signals; and (3) eliminating the influence of respiratory signal modulation on parameter extraction.<<ETX>>

Adaptive filter design for estimation and detection of biological signals: an application to electroencephalography

An innovative technique for estimation and detection of alpha waves using adaptive filters is presented. In these analyses both Wiener transfer functions and non-Wiener transfer functions are used. Comparisons are made for different RMS signal-to-noise ratios. Simulations are done with class 1 filters where prior knowledge of all spectra estimates is assumed. These techniques are found to be very flexible in their application to other biological signals.<<ETX>>This paper presents an innovative technique for estimation and detection of alpha waves using adaptive filters. In these analyses both Wiener transfer functions and non-Wiener transfer functions were used. Comparisons were made for different SNRrms. These techniques were found to be very flexible in their application to other biological signals.

Adaptive filter design for estimation and detection of biological signals: An application to electroencephalography

This paper presents an innovative technique for estimation and detection of alpha waves using adaptive filters. In these analyses both Wiener transfer functions and non-Wiener transfer functions were used. Comparisons were made for different SNR rms . These techniques were found to be very flexible in their application to other biological signals.

Influence Of Signal Fidelity On Impedance Cardiographically Derived Values At Resting And Accelerated Heart Rates

Since the spectrum of the impedance cardiogram (ICG) extends from DC to 50 Hz, any amplifier with an upper band limit less than 50 Hz can be expected to produce attenuation and distortion of the ICG. This Signal attenuation may be systematically enhanced under conditions of high heart rates (HR) when a greater proportion of Signal energy will be In the upper frequency range of the ICG spectrum. Therefore, the present study was designed to assess the influence of amplifier bandwidth and Signal fidelity on dZldtmax, stroke volume (SV), and systolic lime intervals (LVET, PEP, OZ, HI). The performance of commonly available commercial systems was tested over a broad range of HRs. The results demonstrated that a digitally differentiated dZldt signal using a differentlator with a corner frequency of 50 Hz, when compared with the 15 Hz corner frequency used In the commercial Impedance cardiograph, systematically enhanced the dZldtrnax amplitude and SV measurements as HR increased. For SV the increase ranged from 17 to 30% as HR increased from 70 to 150 bpm. Moreover, the digitally filtered signal had greater resolution and produced less prolonged PEP and az intervals and greater HI with increasing HR. These findings Indicate that impedance cardiographs with Insufficient upper band limits will differentially Influence ICG-derived measurements as HR varies.