Bor-Shyh Lin

RTWPMS: A Real-Time Wireless Physiological Monitoring System

This paper demonstrates the design and implementation of a real-time wireless physiological monitoring system for nursing centers, whose function is to monitor online the physiological status of aged patients via wireless communication channel and wired local area network. The collected data, such as body temperature, blood pressure, and heart rate, can then be stored in the computer of a network management center to facilitate the medical staff in a nursing center to monitor in real time or analyze in batch mode the physiological changes of the patients under observation. Our proposed system is bidirectional, has low power consumption, is cost effective, is modular designed, has the capability of operating independently, and can be used to improve the service quality and reduce the workload of the staff in a nursing center

WHEEZE RECOGNITION BASED ON 2D BILATERAL FILTERING OF SPECTROGRAM

This paper describes the design of a low-cost and high performance wheeze recognition system. First, respiratory sounds are captured, amplified and filtered by an analog circuit; then digitized through a PC soundcard, and recorded in accordance with the Computerized Respiratory Sound Analysis (CORSA) standards. Since the proposed wheeze detection algorithm is based on the spectrogram processing of respiratory sounds, spectrograms generated from recorded sounds have to pass through a 2D bilateral filter for edge-preserving smoothing. Finally, the processed spectra go through an edge detection procedure to recognize wheeze sounds. Experiment results show a high sensitivity of 0.967 and a specificity of 0.909 in qualitative analysis of wheeze recognition. Due to its high efficiency, great performance and easy-to-implement features, this wheeze recognition system could be of interest in the clinical monitoring of asthma patients and the study of physiological mechanisms in the respiratory airways.

Higher-Order-Statistics-Based Radial Basis Function Networks for Signal Enhancement

In this paper, a higher-order-statistics (HOS)-based radial basis function (RBF) network for signal enhancement is introduced. In the proposed scheme, higher order cumulants of the reference signal were used as the input of HOS-based RBF. An HOS-based supervised learning algorithm, with mean square error obtained from higher order cumulants of the desired input and the system output as the learning criterion, was used to adapt weights. The motivation is that the HOS can effectively suppress Gaussian and symmetrically distributed non-Gaussian noise. The influence of a Gaussian noise on the input of HOS-based RBF and the HOS-based learning algorithm can be mitigated. Simulated results indicate that HOS-based RBF can provide better performance for signal enhancement under different noise levels, and its performance is insensitive to the selection of learning rates. Moreover, the efficiency of HOS-based RBF under the nonstationary Gaussian noise is stable

A Functional Link Network With Higher Order Statistics for Signal Enhancement

A functional link network with higher order statistics is introduced for signal enhancement. The proposed scheme uses the mean-square error (MSE) between higher order statistics of desired signals and filtered output as the learning criterion for training weights in the functional link network. This is motivated by the fact that higher order statistics have a natural tolerance to Gaussian and symmetrically distributed non-Gaussian noises. Results show that the performance of functional link network with higher order statistics is less sensitive to the selection of learning rates than the conventional functional link network and adaptive line enhancement. It is also demonstrated that it can enhance signal more effectively under different noise levels for stationary and nonstationary Gaussian noises

Ambient noise canceller in pulmonary sound using WHT transform domain adaptive filter

In the process of signal processing of pulmonary sounds, the cancellation of ambient noise is very important. Because the ambient noise is a wide band signal in the frequency domain, it usually uses an adaptive noise canceller (ANC) structure to cancel the ambient noise. However, the usually used algorithm, normalized least mean square (nLMS) algorithm, may fail when dealing with the ambient noise due to the time-varying system because of its low convergence speed. We use a transform domain adaptive filter (TDAF) with a Walsh-Hadamard transform (WHT) to improve the convergence speed. The simulation results show that this structure would cancel the ambient noise more efficiently.

Removing residual power-line interference using WHT adaptive filter

Power-line interference is a common phenomenom in low-frequency biophysical measurement. The usual way of solving this is the use of a fixed bandwidth in an analog or digital notch filter. However, these methods are not very suitable when the power-line interference frequency is non-stationary. In this paper, an effective adaptive filter (ADF) structure is proposed to minimize the residual power-line interference without loss of reality. In order to obtain a satisfactory and acceptable convergence performance, the Walsh-Hadamard (WHT) transform is used in the ADF. Throughout many clinical measurements, the result of this structure is effective in eliminating EMI/EMC interference. However, to overcome the continuous changing frequency of power-line interference and to obtain a better convergence rate, a transform domain adaptive filter (TDADF) is used. An ECG result is shown after removing the power-line interference and its FFT.

Higher Order Statistics-Based Radial Basis Function Network for Evoked Potentials

In this study, higher order statistics-based radial basis function network (RBF) was proposed for evoked potentials (EPs). EPs provide useful information on diagnosis of the nervous system. They are time-varying signals typically buried in ongoing EEG, and have to be extracted by special methods. RBF with least mean square (LMS) algorithm is an effective method to extract EPs. However, using LMS algorithm usually encounters gradient noise amplification problem, i.e., its performance is sensitive to the selection of step sizes and additional noise. Higher order statistics technique, which can effectively suppress Gaussian and symmetrically distributed non-Gaussian noises, was used to reduce gradient noise amplification problem on adaptation in this study. Simulations and human experiments were also carried out in this study.

Adaptive interference cancel filter for evoked potential using high-order cumulants

This paper is to present evoked potential (EP) processing using adaptive interference cancel (AIC) filter with second and high order cumulants. In conventional ensemble averaging method, people have to conduct repetitively experiments to record the required data. Recently, the use of AIC structure with second statistics in processing EP has proved more efficiency than traditional averaging method, but it is sensitive to both of the reference signal statistics and the choice of step size. Thus, we proposed higher order statistics-based AIC method to improve these disadvantages. This study was experimented in somatosensory EP corrupted with EEG. Gradient type algorithm is used in AIC method. Comparisons with AIC filter on second, third, fourth order statistics are also presented in this paper. We observed that AIC filter with third order statistics has better convergent performance for EP processing and is not sensitive to the selection of step size and reference input.

Power-line Interference Removal of Bioelectric Signal Measurement by using Genetic Adaptive Filter

In this study, we proposed a genetic adaptive filter to removing power-line interference. In previous work, the proposed structure, which extracts the interference component from the input biomedical signal to be a reference signal of the adaptive filter to estimate power-line interference, is effective for removing interference. Since this adaptive filter with least-mean square algorithm is sensitive to the eigenvalue spread of the autocorrelation matrix of the reference signals, and the selection of its step-size. Thus, we employ the genetic parallel search technique to improve the least-mean square algorithm. Through simulations, it shows that the adaptive filter with genetic algorithm provides better performance with any step-size

Evoked Potentials Estimation by using Higher Order Adaptive Neural Network filter

Evoked potentials are usually embedded in the ongoing electroencephalogram with a very low signal-to-noise ratio. The neural network filtering technique which has the advantage of complex mapping is one of the applicable methods for evoked potentials estimation. The backpropagation algorithm based on second order statistics is commonly used to adapt neural network filters. However it is easily influenced by additive Gaussian noise. In this study, a neural network filter with a modified back-propagation algorithm for higher order statistics was proposed. With higher-order statistics technique, additive Gaussian noise is suppressed to improve the performance of evoked potentials estimation