Wavelet-based Segmentation and Feature Extraction of Heart Sounds for Intelligent PDA-based Phonocardiography

Objectives : Many pathological conditions of the cardiovascular system cause murmurs and aberrations in heart sounds. Phonocardiography provides the clinician with a complementary tool to record the heart sounds heard during auscultation. The advancement of intracardiac phonocardiography combined with modern digital signal processing techniques has strongly renewed researchers' interest in studying heart sounds and murmurs.

The aim of this work is to investigate the applicability of different spectral analysis methods to heart sound signals and explore their suitability for PDA-based implementation.

Methods : Fourier transform (FT), short-time Fourier transform (STFT) and wavelet transform (WT) are used to perform spectral analysis on heart sounds. A segmentation algorithm based on Shannon energy is used to differentiate between first and second heartsounds. Then wavelet transform is deployed again to extract 64 features of heart sounds.

Results : The FT provides valuable frequency information but the timing information is lost during the transformation process. The STFT or spectrogram provides valuable time-frequency information but there is a trade-off between time and frequency resolution. Waveletanalysis, however, does not suffer from limitations of the STFT and provides adequate time and frequency resolution to accurately characterize the normal and pathological heartsounds.

Conclusions : The results show that the wavelet-based segmentation algorithm is quite effective in localizing the important components of both normal and abnormal heart sounds. They also demonstrate that wavelet-based feature extraction provides suitable feature vectors which are clearly differentiable and useful for automatic classification of heart sounds.

Design and prototyping of a wristband-type wireless photoplethysmographic device for heart rate variability signal analysis

Heart Rate Variability (HRV) signal analysis provides a quantitative marker of the Autonomic Nervous System (ANS) function. A wristband-type wireless photoplethysmographic (PPG) device was custom-designed to collect and analyze the arterial pulse in the wrist. The proposed device is comprised of an optical sensor to monitor arterial pulse, a signal conditioning unit to filter and amplify the analog PPG signal, a microcontroller to digitize the analog PPG signal, and a Bluetooth module to transfer the data to a smart device. This paper proposes a novel model to represent the PPG signal as the summation of two Gaussian functions. The paper concludes with a verification procedure for HRV signal analysis during sedentary activities.

Rapid Prototyping of a Smart Device-based Wireless Reflectance Photoplethysmograph

This paper presents the design, fabrication, and testing of a wireless heart rate (HR) monitoring device based on photoplethysmography (PPG) and smart devices. PPG sensors use infrared (IR) light to obtain vital information to assess cardiac health and other physiologic conditions. The PPG data that are transferred to a computer undergo further processing to derive the Heart Rate Variability (HRV) signal, which is analyzed to generate quantitative markers of the Autonomic Nervous System (ANS). The HRV signal has numerous monitoring and diagnostic applications. To this end, wireless connectivity plays an important role in such biomedical instruments. The photoplethysmograph consists of an optical sensor to detect the changes in the light intensity reflected from the illuminated tissue, a signal conditioning unit to prepare the reflected light for further signal conditioning through amplification and filtering, a low-power microcontroller to control and digitize the analog PPG signal, and a Bluetooth module to transmit the digital data to a Bluetooth-based smart device such as a tablet. An Android app is then used to enable the smart device to acquire and digitally display the received analog PPG signal in real-time on the smart device. This article is concluded with the prototyping of the wireless PPG followed by the verification procedures of the PPG and HRV signals acquired in a laboratory environment.

Wavelet Transform-Based ECG Baseline Drift Removal for Body Surface Potential Mapping

This paper gives a new approach for the removal of slow baseline drift components of electrocardiographic (ECG) signals based on the discrete wavelet transform. The baseline drift is efficiently removed by zeroing the scaling coefficients of the discrete wavelet transform. Such approach can easily be combined with other wavelet based approaches for random noise reduction or power line interference reduction. The new pre-processing approach can remove the low-frequency components without introducing distortions in the ECG waveform.

Evaluation of respiratory system models based on parameter estimates from impulse oscillometry data

Impulse oscillometry offers advantages over spirometry because it requires minimal patient cooperation, it yields pulmonary function data in a form that is readily amenable to engineering analysis. In particular, the data can be used to obtain parameter estimates for electric circuit-based models of the respiratory system, which in turn may assist the detection and diagnosis of various diseases/pathologies. Of the six models analyzed during this study, Mead's model seems to provide the most robust and accurate parameter estimates for our data set of 5 subjects with airflow obstruction including asthma and chronic obstructive pulmonary disease and another 5 normal subjects with no identifiable respiratory disease. Such a diagnostic approach, relying on estimated parameter values from a respiratory system model estimate and the degree of their deviation from the normal range, may require additional measures to ensure proper identification of diseases/pathologies but the preliminary results are promising.

A comparison of linear respiratory system models based on parameter estimates from PRN forced oscillation data

The forced oscillation technique offers some advantages over spirometry for assessing pulmonary function. It requires only passive patient cooperation; it also provides data in a form, frequency-dependent impedance, which is very amenable to engineering analysis. In particular, the data can be used to obtain parameter estimates for electric circuit-based models of the respiratory system, which can in turn aid the detection and diagnosis of various diseases/pathologies. In this study, we compare the least-squares error performance of the RIC, extended RIC, augmented RIC, augmented RIC+I(p), DuBois, Nagels and Mead models in fitting 3 sets of impedance data. These data were obtained by pseudorandom noise forced oscillation of healthy subjects, mild asthmatics and more severe asthmatics. We found that the aRIC+I(p) and DuBois models yielded the lowest fitting errors (for the healthy subjects group and the 2 asthmatic patient groups, respectively) without also producing unphysiologically large component estimates.

Modeling human respiratory impedance in Hispanic asthmatic children

Central (large airway) and peripheral (small airway) dysfunction frequently occur in patients with asthma and chronic obstructive lung disease. Measurement of the respiratory impedance can assist with diagnosis of pathological conditions. The forced Oscillation technique (FOT) superimposes small pressure perturbations at the mouth during tidal breathing of a subject to measure lung mechanical parameters. The Impulse Oscillometry System (IOS) is a commercial instrument that measures forced oscillatory impedance. IOS can be conveniently used in children as it only requires their passive cooperation during pulmonary function testing. Forced oscillatory impedance can be analyzed with respiratory system equivalent electrical circuit models. Models of varying complexity and fidelity have been developed to provide better understanding of respiratory mechanics and enable greater specificity of the diagnosis. Parameter estimates for these models can be used as reference values for detection and diagnosis of different respiratory pathologies. Previous work by our group has evaluated several known respiratory models and a new RIC model (augmented RIC) has emerged which offers advantages over earlier models. It has been shown that one parameter of this new model (representing peripheral airway compliance) is capable of discriminating between normal and asthmatic children. In this paper, we analyzed IOS data from 40 Hispanic asthmatic children and obtained sensitive impulse oscillometric parameters of lung function as well as parameter estimates for the augmented RIC (aRIC) model to distinguish between constricted (asthmatic condition) and non-constricted (non-asthmatic condition) airways with very promising results.

Comparison of heart rate variability signal features derived from electrocardiography and photoplethysmography in healthy individuals

The heart rate variability (HRV) signal is indicative of autonomic regulation of the heart rate (HR). It could be used as a noninvasive marker in monitoring the physiological state of an individual. Currently, the primary method of deriving the HRV signal is to acquire the electrocardiogram (ECG) signal, apply appropriate QRS detection algorithms to locate the R wave and its peak, find the RR intervals, and perform suitable interpolation and resampling to produce a uniformly sampled tachogram. This process could sometimes result in errors in the HRV signal due to drift, electromagnetic and biologic interference, and the complex morphology of the ECG signal. The photoplethysmographic (PPG) signal has the potential to eliminate the problems with the ECG signal to derive the HRV signal. To investigate this point, a PDA-based system was developed to simultaneously record ECG and PPG signals to facilitate accurately controlled sampling and recording durations. Two healthy young volunteers participated in this pilot study to evaluate the applicability of our approach. To improve data quality, ECG and PPG recordings were acquired three times/subject. A comparison between different features of the HRV signals derived from both methods was performed to test the validity of using PPG signals in HRV analysis. We used autoregressive (AR) modeling, Poincare' plots, cross correlation, standard deviation, arithmetic mean, skewness, kurtosis, and approximate entropy (ApEn) to derive and compare different measures from both ECG and PPG signals. This study demonstrated that our PDA-based system was a convenient and reliable means for acquisition of PPG-derived and ECG-derived HRV signals. The excellent agreement between different measures of HRV signals acquired from both methods provides potential support for the idea of using PPGs instead of ECGs in HRV signal derivation and analysis in ambulatory cardiac monitoring of healthy individuals.

Assessment of Itakura Distance as a valuable feature for computer-aided classification of sleep stages

Staging and detection of various states of sleep derived from EEG and other biomedical signals have proven to be very helpful in diagnosis, prognosis and remedy of various sleep related disorders. The time consuming and costly process of visual scoring of sleep stages by a specialist has always motivated researchers to develop an automatic sleep scoring system and the first step toward achieving this task is finding discriminating characteristics (or features) for each stage. A vast variety of these features and methods have been investigated in the sleep literature with different degrees of success. In this study, we investigated the performance of a newly introduced measure: the Itakura Distance (ID), as a similarity measure between EEG and EOG signals. This work demonstrated and further confirmed the outcomes of our previous research that the Itakura Distance serves as a valuable similarity measure to differentiate between different sleep stages.

A study of IOS data using two mead-related models of respiratory impedance

This paper introduces two new respiratory system models, the Mead-Cw model and the Mead-Cl model, which are 6-component models that are intermediate in complexity between the well-known 7-component Mead model and the recently proposed 5-component augmented RIC model (derived from the Mead model by eliminating both Cw and Cl). Their modeling errors were compared to the RIC, extended RIC, augmented RIC and Mead models, for component values estimated from IOS data. The two new models yielded lower errors than all the other models, except for the Mead model. However, the Mead-Cl model and the Mead-Cw model also yielded unreasonably large values for Cw and Cl, respectively, which are known disadvantages of the Mead model. Hence the augmented RIC model appears to be the most useful at present for IOS-based computer-aided detection and diagnosis of respiratory disorders.

EOG and EMG: two important switches in automatic sleep stage classification

Sleep is a natural periodic state of rest for the body, in which the eyes are usually closed and consciousness is completely or partially lost. In this investigation we used the EOG and EMG signals acquired from 10 patients undergoing overnight polysomnography with their sleep stages determined by expert sleep specialists based on RK rules. Differentiation between Stage 1, Awake and REM stages challenged a well trained neural network classifier to distinguish between classes when only EEG-derived signal features were used. To meet this challenge and improve the classification rate, extra features extracted from EOG and EMG signals were fed to the classifier. In this study, two simple feature extraction algorithms were applied to EOG and EMG signals. The statistics of the results were calculated and displayed in an easy to visualize fashion to observe tendencies for each sleep stage. Inclusion of these features show a great promise to improve the classification rate towards the target rate of 100%

Can asthma in children be detected by the estimated parameter values of the augmented RIC model?

This paper describes the estimation of the parameter values for the recently introduced augmented RIC respiratory system model from impulse oscillometry data obtained from both asthmatic and normal children. An analysis of these values has indicated that one of the capacitance parameters of the model provides good discrimination between these two groups of children; moreover, this finding corresponds well with current medical understanding of the pathology of asthma.

Nonlinear dynamics analysis of heart rate variability signals to detect sleep disordered breathing in children

This paper reports a preliminary investigation to evaluate the significance of various nonlinear dynamics approaches to analyze the heart rate variability (HRV) signal in children with sleep disordered breathing (SDB). Data collected from children in the age group of 1-17 years diagnosed with sleep apnea were used in this study. Both short term (5 minutes) and long term data from a full night polysomnography (7-9 hours) were analyzed. For short term data, the presence of nonstationarity in the derived HRV signal was determined by calculating the local Hurst exponent. Poincare plots and approximate entropy (ApEn) were then used to show the presence of correlation in the data. For long term data, the derived HRV signal was first separated into corresponding sleep stages with the aid of the recorded sleep hypnogram values at 30 seconds epochs. The scaling exponents using detrended fluctuation analysis (DFA) and the ApEn were then calculated for each sleep stage. Data from two sample subjects recorded for different sleep stages and breathing patterns were considered for short term analysis. Data from 7 sample subjects (after sleep staging) were considered for long term analysis. The accuracy rate of ApEn was about 72% for both long term and short term data sets. The accuracy rate of Alpha (alpha) derived from DFA for long term correlations was 57%. Further work is necessary to improve on the accuracies of these useful nonlinear dynamic measures and determine their sensitivity and specificity to detect SDB in children.

A comparison of various respiratory system models based on parameter estimates from impulse oscillometry data

Impulse oscillometry offers an advantage over spirometry when conducting pulmonary function tests. Not only does it require minimal patient cooperation, it provides useful data in a form amenable to engineering methods. In particular, the data can be used to obtain parameter estimates for electric circuit-based models of the respiratory system, which can in turn aid the detection and diagnosis of various diseases/pathologies. Of the six models analyzed during this study, the DuBois model and a newly proposed extended RIC model seem to provide the most robust parameter estimates for our entire data set of 106 subjects with various respiratory ailments such as asthma and chronic obstructive pulmonary disease. Such a diagnostic approach, relying on estimated parameter values, may require additional measures to ensure proper identification of diseases/pathologies but the preliminary results are promising.

EEG feature extraction for classification of sleep stages

Automated sleep staging based on EEG signal analysis provides an important quantitative tool to assist neurologists and sleep specialists in the diagnosis and monitoring of sleep disorders as well as evaluation of treatment efficacy. A complete visual inspection of the EEG recordings acquired during nocturnal polysomnography is time consuming, expensive, and often subjective. Therefore, feature extraction is implemented as an essential preprocessing step to achieve significant data reduction and to determine informative measures for automatic sleep staging. However, the analysis of the EEG signal and extraction of sensitive measures from it has been a challenging task due to the complexity and variability of this signal. We present three different schemes to extract features from the EEG signal: relative spectral band energy, harmonic parameters, and Itakura distance. Spectral estimation is performed by using autoregressive (AR) modeling. We then compare the performance of these schemes with the view to select an optimal set of features for specific, sensitive, and accurate neuro-fuzzy classification of sleep stages.

Detrended Fluctuation Analysis: A Suitable Long-term Measure of HRV Signals in Children with Sleep Disordered Breathing

On the body surface the electric field generated by the cardiac muscles consists of electric potential maxima and minima that increase and decrease during each cardiac cycle. The recording of these electric potentials as a function of time is called electrocardiography, and the resulting signal is called the electrocardiogram (ECG). The ECG signal is used extensively as a low cost diagnostic tool to provide information concerning the heart's state of health. Reliable and accurate detection of the QRS complex and R wave peak in ECG signals is essential in computer-based ECG analysis. In this paper we evaluate the significance of Detrended Fluctuation Analysis (DFA) for studying heart rate variability in children with sleep disordered breathing. An Enhanced Hilbert Transform (EHT) algorithm was used to derive the Heart Rate Variability (HRV) signal. We compare the DFA values with Approximate Entropy and Poincaré Plots of HRV signals as these are very useful in characterization and visualization of HRV data. Our data demonstrated differences in DFA parameters between periods of normal and abnormal breathing and also between sleep stages. These results suggest that DFA is suitable for the long-term analysis of non-stationary time series such as HRV signals and may also be applied in the detection of sleep disordered breathing.

Itakura Distance: A Useful Similarity Measure between EEG and EOG Signals in Computer-aided Classification of Sleep Stages

Sleep is a natural periodic state of rest for the body, in which the eyes usually close and consciousness is completely or partially lost. Consequently, there is a decrease in bodily movements and responsiveness to external stimuli. Slow wave sleep is of immense interest as it is the most restorative sleep stage during which the body recovers from weariness. During this sleep stage, electroencephalographic (EEG) and electro-oculographic (EOG) signals interfere with each other and they share a temporal similarity. In this investigation we used the EEG and EOG signals acquired from 10 patients undergoing overnight polysomnography with their sleep stages determined by certified sleep specialists based on RK rules. In this pilot study, we performed spectral estimation of EEG signals by Autoregressive (AR) modeling, and then used Itakura Distance to measure the degree of similarity between EEG and EOG signals. We finally calculated the statistics of the results and displayed them in an easy to visualize fashion to observe tendencies for each sleep stage. We found that Itakura Distance is the smallest for sleep stages 3 and 4. We intend to deploy this feature as an important element in automatic classification of sleep stages.

Digitization and synchronization method for electrocardiogram printouts

A method of digitization and synchronization of ECG signals for use in vectorcardiography is hereby presented. It is intended to allow the computerized use of the ECG printout used by most cardiologists. The problem is solved by digitizing the printout, which can then be used for vectorcardiography and other modern techniques, once it is synchronized using cross-correlation. Cross-correlation was found to be a superior technique for synchronization over other methods, like R-peak synchronization that can be erroneously used, due to time correspondence. The highest and lowest normalized rms error of the digitized and original signals was found to be 13.49% and 9.85% for recording V1 and Lead III, respectively. The error found was expected because of the comparison of an image with many points in a single instant of time, in comparison to only one point as it is in a digital signal. The result obtained from the work presented here is that the ECG printout needs no longer to remain only as a graphical report, but could also be used for computerized analyses.

Accurate derivation of heart rate variability signal for detection of sleep disordered breathing in children

The electrocardiogram (ECG) signal is used extensively as a low cost diagnostic tool to provide information concerning the heart's state of health. Accurate determination of the QRS complex, in particular, reliable detection of the R wave peak, is essential in computer based ECG analysis. ECG data from Physionet's Sleep-Apnea database were used to develop, test, and validate a robust heart rate variability (HRV) signal derivation algorithm. The HRV signal was derived from pre-processed ECG signals by developing an enhanced Hilbert transform (EHT) algorithm with built-in missing beat detection capability for reliable QRS detection. The performance of the EHT algorithm was then compared against that of a popular Hilbert transform-based (HT) QRS detection algorithm. Autoregressive (AR) modeling of the HRV power spectrum for both EHT- and HT-derived HRV signals was achieved and different parameters from their power spectra as well as approximate entropy were derived for comparison. Poincare plots were then used as a visualization tool to highlight the detection of the missing beats in the EHT method After validation of the EHT algorithm on ECG data from the Physionet, the algorithm was further tested and validated on a dataset obtained from children undergoing polysomnography for detection of sleep disordered breathing (SDB). Sensitive measures of accurate HRV signals were then derived to be used in detecting and diagnosing sleep disordered breathing in children. All signal processing algorithms were implemented in MATLAB. We present a description of the EHT algorithm and analyze pilot data for eight children undergoing nocturnal polysomnography. The pilot data demonstrated that the EHT method provides an accurate way of deriving the HRV signal and plays an important role in extraction of reliable measures to distinguish between periods of normal and sleep disordered breathing (SDB) in children.

Wavelet-based segmentation and feature extraction of heart sounds for intelligent PDA-based phonocardiography

OBJECTIVES

Many pathological conditions of the cardiovascular system cause murmurs and aberrations in heart sounds. Phonocardiography provides the clinician with a complementary tool to record the heart sounds heard during auscultation. The advancement of intracardiac phonocardiography combined with modern digital signal processing techniques has strongly renewed researchers' interest in studying heart sounds and murmurs. The aim of this work is to investigate the applicability of different spectral analysis methods to heart sound signals and explore their suitability for PDA-based implementation.

METHODS

Fourier transform (FT), short-time Fourier transform (STFT) and wavelet transform (WT) are used to perform spectral analysis on heart sounds. A segmentation algorithm based on Shannon energy is used to differentiate between first and second heart sounds. Then wavelet transform is deployed again to extract 64 features of heart sounds.

RESULTS

The FT provides valuable frequency information but the timing information is lost during the transformation process. The STFT or spectrogram provides valuable time-frequency information but there is a trade-off between time and frequency resolution. Wavelet analysis, however, does not suffer from limitations of the STFT and provides adequate time and frequency resolution to accurately characterize the normal and pathological heart sounds.

CONCLUSIONS

The results show that the wavelet-based segmentation algorithm is quite effective in localizing the important components of both normal and abnormal heart sounds. They also demonstrate that wavelet-based feature extraction provides suitable feature vectors which are clearly differentiable and useful for automatic classification of heart sounds.

Reducing power line interference in digitised electromyogram recordings by spectrum interpolation

Interference from power lines (50 or 60 Hz) is the largest source of extraneous noise in many bio-electric signals and is within the bandwidth of many such signals. In this study, two different methods were compared for their efficacy in removing 50 Hz noise added to surface electromyogram (EMG) signals free of power line interference. The first was a simple second-order recursive digital notch filter. The second was an approach called spectrum interpolation, in which it is assumed that the magnitude of the original 50 Hz component of the EMG signal can be approximated by interpolation of the amplitude spectrum of the signal. When the spectrum was based on records containing an integer number of cycles of 50 Hz interference, and the frequency resolution was finer than 1 Hz, spectrum interpolation performed similarly to, or significantly better than, the notch filter (p < 0.01). It was also possible to make spectrum interpolation more robust than the notch filter. The Pearson squared correlation coefficient r2 between clean signals and signals processed using the notch filter was reduced from 0.98 to 0.65 when the interference frequency was increased by 0.5 Hz, but r2 for spectrum interpolation at 0.2 Hz resolution was only reduced from 0.99 to 0.85 if spectral values between approximately 49.5 and 50.5 Hz were modified by interpolation.

A device for investigating neuromuscular control in the human masticatory system

A new apparatus has been developed to study the control of mastication in humans. The subject places his/her teeth on fixed upper and mobile lower bite plates; the device then enables opening and closing movements of the lower jaw against a controlled resistance. It is also possible to vary the number of teeth in contact with the device during an experiment from the entire dental arcade to a single tooth. The specially designed lower bite plate is dynamic and allows for both rotation and translation of the lower jaw during movement, thus, permitting the natural curvilinear trajectory of the jaw. The lower bite plate can follow chewing initiated by the subject without resisting the movement ('no force' mode) via a dedicated microprocessor controlled compensation mechanism. Another function of the device is to inject a constant predetermined load onto the lower bite plate so that the subject 'chews' against a fixed resistance simulating rapidly yielding food bolus ('fixed force' mode). The device can be programmed to increase or decrease the force during the closing or opening phase of chewing by feeding the position information into the force compensation system so both position and force change in parallel, hence, simulating a bite onto a non-yielding, or sticky, food bolus ('normal chewing' mode). By use of a jaw position compensation mechanism, the device can actively move the lower jaw, following any imposed position pattern ('position controlled' mode). The chewing simulator also has a mode that holds the position at a fixed level and allows the force to change ('position hold' mode). Furthermore, the device can inject additional rapid or slow forces or displacements onto the lower bite plate in order to elicit reflexes so that the response of jaw muscles to such stimuli can be examined at various jaw positions, force levels, phases of motion and velocities. The different modes of the apparatus can be used to study the operation and feedback control of human mastication; in particular whether modulations in jaw muscle activity and reflexes are due to changes in force, velocity, position, chewing cycle phase or a combination of these factors.

Compensation of limb weight on interfaced raw torque signals from a KIN-COM® dynamometer to an AMLAB® Workstation

The effect of gravity should be considered when using isokinetic devices to measure human movement performance. In most isokinetic dynamometers gravity compensation is controlled by software through a gravity correction option. However in some complex research protocols the dynamometer signal acquisition and processing capability is not adequate to effectively synchronize or process a wide range of captured signals. Therefore when the force/torque signals from a commonly used dynamometer such as KIN-COM are interfaced into a signal processing workstation such as AMLAB, it is necessary to further process the received raw signals for gravity correction. The aim of this study was to evaluate the effectiveness of an AMLAB-based instrument designed for gravity compensation of raw torque signals acquired from a KIN-COM dynamometer. To check the accuracy of weight compensation within the AMLAB, environment, torque signals produced by a known weight during a 180-degree range of KIN-COM lever arm movement were compared with and without weight compensation. The results indicated that this technique is an accurate means for weight compensation when raw torque signals from a KIN-COM dynamometer are interfaced to an AMLAB workstation.

An integrated AMLAB-based system for acquisition, processing and analysis of evoked EMG and mechanical responses of upper limb muscles

An integrated multi-channel AMLAB-based data acquisition, processing and analysis system has been developed to simultaneously display, quantify and correlate electromyographic (EMG) activity, resistive torque, range of motion, and pain responses evoked by passive elbow extension in humans. The system was designed around the AMLAB analog modules and software objects called ICAMs. Each channel consisted of a time and frequency domain block, a torque and angle measurement block, an experiment number counter block and a data storage and retrieval block. The captured data in each channel was used to display and quantify: raw EMG, rectified EMG, smoothed rectified EMG, root-mean-squared EMG, fast Fourier transformed (FFT) EMG, and normalized power spectrum density (NPSD) of EMG. Torque and angle signals representing elbow extension measured by a KIN-COM dynamometer during neural tension testing, as well as signals from an electronic pain threshold marker were interfaced to AMLAB and presented in one integrated display. Although this system has been designed to specifically study the patterns and nature of evoked motor responses during clinical investigation of carpal tunnel syndrome (CTS) patients, it could equally well be modified to allow acquisition, processing and analysis of EMG signals in other studies and applications. In this paper, we present for the first time the steps involved in the design, implementation and testing of an integrated AMLAB-based system to study and analyse the mechanically evoked electromyographic, torque and ROM signals and correlate various levels of pain to these signals. We also present samples of resistive torque ROM, and raw and processed EMG recordings during passive elbow extension.

ECG compression: evaluation of FFT, DCT, and WT performance

This work investigates a set of ECG data compression schemes to compare their performances in compressing and preparing ECG signals for automatic cardiac arrhythmia classification. These schemes are based on transform methods such as fast Fourier transform (FFT), discrete cosine transform (DCT), wavelet transform (WT), and their combinations. Each specific transform is applied to a pre-selected data segment from the MIT-BIH database and then compression is performed in the new domain. These transformation methods are known as an important class of ECG compression techniques. The WT has been shown as the most efficient method for further improvement. A compression ratio of 7.98 to 1 has been achieved with a percent of root mean square difference (PRD) of 0.25%, indicating that the wavelet compression technique offers the best performance over the other evaluated methods.

Power spectrum analysis of human femoral rotations during gait

Femoral rotational waveforms sampled at 15 Hz for a duration 20 s satisfy the sampling rate and frequency resolution requirements of such waveforms during walking. Spectral analysis provides a unique signature of the frequency composition of such signals. This identity may prove useful in characterising human gait and could be of value in future studies of walking in health and disease.

The enteric neural network and three dimensional computer modelling of intestinal peristalsis

A computer model of the enteric nervous system has been developed using MATLAB in order to determine the extent to which the nature of intestinal activity can be explained by our current understanding of the projections and connectivity of enteric neurons. The model is based on repeated, identical overlapping modules, each of which contains the same number of neurones and circular muscle. The connections between modules were derived from microanatomical data. This simple model explains some characteristic features of the generation of an intestinal motor pattern.

Development of a low cost microP-based blood gas monitor

A low cost microprocessor-based system for continuous in-vivo measurement of blood pH and pCO2 using a biosensor is being developed. The biosensor is a pH Ti2O3 ISFET (Ion Sensitive Field Effect Transistor) with a sensitivity of -56 mV/pH and a linear response over physiologically meaningful blood pH range of 6.5 to 8.0. The ISFET chip is bonded onto a printed circuit board substrate to enable reliable and robust connections to three wires. The small sensor tip sized of 0.95 mm facilitates its incorporation into a catheter tip and its placement into a rat's artery. A microcontroller supervises preusage calibration, data logging, display of alphanumeric information on a LCD (liquid crystal display) unit, and resettable visual & audio alarm indicators. Preliminary clinical testing are underway to perfect the system and use it on a routine basis to monitor the physiological state of laboratory animals during neurophysiological experiments.

Biomedical image processing in pathology: a review

Pathologists make a diagnostic decision by viewing a specimen and measuring various diagnostically important attributes of an isolated object such as size, shape, darkness, colour and texture. This is a complex process. In recent years, computer-aided image processing and analysis systems have played a significant role in quantitative pathology. This paper summarises basic image processing and analysis techniques and reviews related work in pathology and cytology based on computational image processing since 1987. Firstly, we present a general introduction to image enhancement, segmentation, morphometry and visualisation for those medical colleagues who may not have the necessary background in this area. (The mathematical treatment is kept to minimum and appropriate references are cited to satisfy the more mathematically oriented readers. Selected examples are provided to demonstrate the effects of various basic image processing algorithms on a MRI scan. It should be emphasised that the reviewed techniques are generally used as preprocessing steps in analysing microscopic images and powerful algorithms are more sophisticated and problem-specific.) Secondly, we review image cytometric and histometric methods, standards, calibration and applications. Finally, we touch upon three dimensional confocal image processing and analysis, applications of artificial neural networks, and optical disk database management for recording and retrieving a large number of digitised high resolution images. The development of integrated optical microscope and computer, systems is also briefly described.