Blind separation of multichannel electrogastrograms using independent component analysis based on a neural network

The electrogastrogram (EGG) is an abdominal surface measurement of gastric myo-electrical activity which regulates gastric contractions. It is of great clinical importance to record and analyse multichannel EGGs, which provide more information on the propagation and co-ordination of gastric contractions. EGGs are, however, contaminated by myo-electric interference from other organs and artefacts such as motion and respiration. The aim of the study is to separate the gastric signal from noisy multichannel EGGs without any information on the interference, using independent component analysis. A neural-network model is proposed, and corresponding unsupervised learning algorithms are developed to achieve the separation. The performance of the proposed method is investigated using artificial data simulating real EGG signals. Experimental EGG data are obtained from humans and dogs. The processed results of both simulated and real EGG data show the following: first, the proposed method is able to separate normal gastric slow waves from respiratory artefacts and random noises. It is also able to extract gastric slow waves, even when the EGG is contaminated by severe respiratory and ECG artefacts. Secondly, when the stomach contains various gastric electric signals with different frequencies, the proposed method is able to separate these different signals, as illustrated by simulations. These data suggest that the proposed method can be used to separate gastric slow waves, respiratory and motion artefacts, and intestinal myo-electric interference that are mixed in the EGG. It can also be used to detect gastric slow-wave uncoupling, during which the stomach has multiple gastric signals with different frequencies. It is believed that the proposed method may also be applicable to other biomedical signals.     

Adaptive independent component analysis of multichannel electrogastrograms.

The electrogastrogram (EGG), a cutaneous measurement of gastric electrical activity, can be severely contaminated by endogenous biological noise sources such as respiratory signal. Therefore it is important to establish effective artifact removal methods. In this paper, a novel blind signal separation method with a flexible non-linearity is introduced and applied to extract the gastric slow wave from multichannel EGGs. Simulation results show that our algorithm is able to separate a wide range of source signals, including mixtures of Gaussian sources. On real data, we demonstrate the successful applications of our procedure to extract the gastric slow wave from multichannel EGGs. As a result, the extracted clean gastric slow wave can be used to facilitate further analysis, e.g. as a reference signal for multichannel adaptive enhancement of the EGG.     

Electrogastrography: measuremnt,analysis and prospective applications

Electrogastrography (EGG), the cutaneous recording of the myoelectrical activity of the stomach using surface electrodes, is attractive due to its non-invasiveness and its correlation with gastric motility. Since the first measurement of the EGG, a great deal of research effort has been expended on investigating its relationship with specific activities of the stomach in question. In this review, we discuss how to measure the EGG, what information is available in the EGG, how to extract useful information from the EGG and the prospective applications of the EGG. After a brief historic evolution of the EGG, we first discuss the measurement of the EGG, the importance of localisation of the stomach and the characteristics of the EGG. The most commonly used EGG data-analysis methods, both power spectral analysis and waveform analysis, are then outlined. Some of prospective applications of the EGG in clinical diagnosis and medical research are described, mainly concerning the correlation between EGG amplitude and gastric motility and that between EGG rhythmic variations and motility disorders. Specifically, we present studies on pre- and postprandial EGGs, observation of the propagation direction of the gastric slow wave and gastric electrical dysrhythmias and their correlation with gastric motility.     

Motor and electrogastrographic activity of the gastric tube formed after esophagectomy.

In order to characterize the motor activity of a surgically constructed gastric tube, several hours of ambulatory intraluminal pressure recordings were performed in 6 patients following esophagectomy and gastric tube construction. Whole pressure waves were spectrally analyzed by Fast Fourier Transform (FFT). Simultaneous abdominal and thoracic electrogastrograms (EGGs) were recorded for about 20 min both before and after meals during ambulatory pressure recording. The pressure waves and EGGs for each 20 min recording were analyzed by the maximal entropy method (MEM). While the motility index of the pressure waves decreased after a meal, the 3 cpm component of these waves (2.4-3.7 cpm) increased significantly (n=6, P<0.05). Both bradygastria (0-2.4 cpm) and the duodeno-respiratory component (10-15 cpm) decreased, while the tachygastria component (3.7-10 cpm) increased, although these differences were not significant. The peak power of the gastric tube abdominal EGGs was significantly larger than that of control abdominal or thoracic EGGs in each of the 1 cpm (0-2.4 cpm), 6 cpm (5.0-7.4) and 8 cpm components (7.5-9.9). The thoracic EGG consisted mainly of the 3 cpm component, while the spectral amplitudes of the 1, 6, 8 and 10 cpm components were below 6% of the 3 cpm component. The peak spectral frequency both of the intraluminal pressure waves by FFT and of the thoracic EGGs by MEM occurred within the 3 cpm component. A cross correlation of about 0.2-0.3 occurred between the thoracic EGGs and the intraluminal pressure waves. Thus the gastric tube seems to preserve most of the original gastric motor characteristics and to contribute as a substitute for the original esophagus and stomach.     

Reconstruction of gastric slow wave from finger photoplethysmographic signal using radial basis function neural network

Extraction of extra-cardiac information from photoplethysmography (PPG) signal is a challenging research problem with significant clinical applications. In this study, radial basis function neural network (RBFNN) is used to reconstruct the gastric myoelectric activity (GMA) slow wave from finger PPG signal. Finger PPG and GMA (measured using Electrogastrogram, EGG) signals were acquired simultaneously at the sampling rate of 100 Hz from ten healthy subjects. Discrete wavelet transform (DWT) was used to extract slow wave (0–0.1953 Hz) component from the finger PPG signal; this slow wave PPG was used to reconstruct EGG. A RBFNN is trained on signals obtained from six subjects in both fasting and postprandial conditions. The trained network is tested on data obtained from the remaining four subjects. In the earlier study, we have shown the presence of GMA information in finger PPG signal using DWT and cross-correlation method. In this study, we explicitly reconstruct gastric slow wave from finger PPG signal by the proposed RBFNN-based method. It was found that the network-reconstructed slow wave provided significantly higher (P < 0.0001) correlation (≥0.9) with the subject’s EGG slow wave than the correlation obtained (≈0.7) between the PPG slow wave from DWT and the EEG slow wave. Our results showed that a simple finger PPG signal can be used to reconstruct gastric slow wave using RBFNN method.     

体表胃电图慢波提取方法的比较研究

体表胃电图(electrogastrogram,EGG)具有幅值小、频率低和窄带宽的特点,并且极易受到心电、呼吸以及运动等干扰.文中对三种常用的EGG慢波的提取方法进行了比较,包括自适应滤波、小波多尺度分解和经验模态分解,并将三种方法分别应用于模拟的含噪EGG信号以及临床EGG慢波的提取.研究结果表明,三种方法各有优缺点,均能有效地提取出EGG信号中的慢波成分.与自适应滤波相比,小波多分辨率分解和经验模态分解获得了更高的信噪比和更低的重构方差.     

Computer simulation of the impact of different dimensions of the stomach on the validity of electrogastrograms

The impact of the dimensions of the stomach on cutaneous recordings of gastric electrical activity (GEA) has not been adequately studied. The stomach was represented as a truncated conoid in a spherical coordinate system. The gastric electric field was modelled using a previously described methodology. Electrical potentials were calculated from sets of points simulating standard cutaneous recordings. The frequency of the signals was maintained at 3 cycles min⁻¹ (period of repetition: 20s), while the velocity of propagation of the depolarisation waves was reduced relative to the reduction in gastric dimensions. The signals were digitally contaminated with a random artificial artefact with a constant amplitude range of 0.2 mV, while the dimensions of the conoid (the circumferential radii and the length of the central axis) were decreased by factors of 1.5, two, four, six and eight. Simulated EGG signals were evaluated quantitatively. Simulated EGG records contaminated with random signals recorded from stomachs with decreasing dimensions exhibited non-linearly increasing standard deviations (p<0.001). Randomly contaminated EGGs calculated from stomachs with dimensions reduced four, six and eight times were abnormal according to previously established quantitative criteria. Computer modelling indicated that gastric dimensions can significantly influence the validity of EGGs. These findings could be particularly important in a paediatric electrogastrography.     

Application of combined genetic algorithms with cascade correlation to diagnosis of delayed gastric emptying from electrogastrograms

The current standard method (radioscintigraphy) for the diagnosis of delayed gastric emptying (GE) of a solid meal involves radiation exposure and considerable expense. Based on combining genetic algorithms with the cascade correlation learning architecture, a neural network approach is proposed for the diagnosis of delayed GE from electrogastrograms (EGGs). EGGs were measured by placing surface electrodes on the abdominal skin over the stomach in 152 patients with suspected gastric motility disorders for 30 min in the fasting state and for 2 h after a standard test meal. The GE rate of the stomach was simultaneously monitored after the meal using radioscintigraphy. Five spectral parameters of EGG data in each patient were used as the inputs to a classifier. The classifier was designed by using genetic algorithms in conjunction with the cascade correlation learning architecture. The main advantage of this technique over the back-propagation (BP) for supervised learning is that it can automatically develop the architecture of neural networks to give a suitable network size for a specific problem. The resulted neural network with three hidden units exhibits 83% correct classification for the EGG data, and has comparable performance with the BP network. This study demonstrates the potential of the neural network approach based on combined genetic algorithms with cascade correlation for diagnosis of gastric emptying from the EGG.     

Characterization of Gastric Electrical Activity Using Magnetic Field Measurements: A Simulation Study

Gastric disorders are often associated with abnormal propagation of gastric electrical activity (GEA). The identification of clinically relevant parameters of GEA using noninvasive measures would therefore be highly beneficial for clinical diagnosis. While magnetogastrograms (MGG) are known to provide a noninvasive representation of GEA, standard methods for their analysis are limited. It has previously been shown in simplistic conditions that the surface current density (SCD) calculated from multichannel MGG measurements provides an estimate of the gastric source location and propagation velocity. We examine the accuracy of this technique using more realistic source models and an anatomically realistic volume conductor model. The results showed that the SCD method was able to resolve the GEA parameters more reliably when the dipole source was located within 100 mm of the sensor. Therefore, the theoretical accuracy of SCD method would be relatively diminished for patients with a larger body habitus, and particularly in those patients with significant truncal obesity. However, many patients with gastric motility disorders are relatively thin due to food intolerance, meaning that the majority of the population of gastric motility patients could benefit from the methods developed here. Large errors resulted when the source was located deep within the body due to the distorting effects of the secondary sources on the magnetic fields. Larger errors also resulted when the dipole was oriented normal to the sensor plane. This was believed to be due to the relatively small contribution of the dipole source when compared to the field produced by the volume conductor. The use of three orthogonal magnetic field components rather than just one component to calculate the SCD yielded marginally more accurate results when using a realistic dipole source. However, this slight increase in accuracy may not warrant the use of more complex vector channels in future superconducting quantum interference device designs. When multiple slow waves were present in the stomach, the SCD map contained only one maximum point corresponding to the more dominant source located in the distal stomach. Parameters corresponding to the slow wave in the proximal stomach were obtained once the dominant slow terminated at the antrum. Additional validation studies are warranted to address the utility of the SCD method to resolve parameters related to gastric slow waves in a clinical setting.     

Continuous Wavelet Analysis as an Aid in the Representation and Interpretation of Electrogastrographic Signals

This paper presents an application of the continuous wavelet transform (CWT) in the analysis of electrogastrographic (EGG) signals. Due to the nonstationary nature of EGG signals, the CWT method, which uses multiresolution scaled windows, gives a better time-frequency resolution than the short-time Fourier transform, which uses a fixed window. Spike activity due to gastric contraction was investigated through experiments on dogs. During spike activity we observed an increase in magnitude of the slow wave and the appearance of a low frequency component with half the frequency of the slow wave. Studies of the EGG signals from the small intestine are also presented to investigate the hypothesis that its slow wave might be confounded with spike activity in the stomach due to the similarity of their frequency ranges. © 1998 Biomedical Engineering Society. PAC98: 0230-f, 8759Wc     

Spatial filters based on independent component analysis for magnetic noise reduction in the magnetocardiogram

A spatial filter design method to reduce magnetic noise in the magnetocardiogram (MCG) is introduced. Based on the facts that external magnetic noise appearing on multichannel MCG sensors is independent of the cardiac signals and that there is strong spatial correlation among the channels, the independent component analysis (ICA) method was applied to extract the noise components from the measured MCG signals. After extraction of the noise components in a given time period using ICA, a spatial filter was made to reduce the noise components in subsequently acquired MCG signals. In experimental studies of nine healthy volunteers, the spatial filters improved the signal-to-noise ratio of the MCG signals by about 500% on average. This spatial filtering method can be used for measurements of MCG signals in a magnetically noisy environment.     

一种韧性的胃电信号传导速率估计方法

由于棘波和/或动作伪迹等因素的影响,实际测得的胃电信号常常具有脉冲性。根据分数低阶统计量理论,本研究提出一种基于最小平均p范数盲信道辨识的韧性时间延迟估计算法(BCILMP),对胃电信号传导速率进行估计。与最小均方误差时间延迟估计算法(LMSTDE)比较,该算法在高斯和脉冲噪声环境下,均可以较好地估计出时间延迟,计算机仿真验证了其性能。运用该方法对四位胃轻瘫病人的胃电信号进行了传导速率的估计。结果表明:虽然胃轻瘫病人的胃慢波也由胃体大弯侧上1/3处向下端幽门处传播,但是其胃慢波的平均传导速率均比正常人的慢,速率的变化也不都是单调增加的。传导是胃电活动最重要的空间特征之一,时间延迟估计方法可以定量地评价胃慢波传导的速率和模式。     

Torso-Tank Validation of High-Resolution Electrogastrography (EGG): Forward Modelling,Methodology and Results

Electrogastrography (EGG) is a non-invasive method for measuring gastric electrical activity. Recent simulation studies have attempted to extend the current clinical utility of the EGG, in particular by providing a theoretical framework for distinguishing specific gastric slow wave dysrhythmias. In this paper we implement an experimental setup called a ‘torso-tank’ with the aim of expanding and experimentally validating these previous simulations. The torso-tank was developed using an adult male torso phantom with 190 electrodes embedded throughout the torso. The gastric slow waves were reproduced using an artificial current source capable of producing 3D electrical fields. Multiple gastric dysrhythmias were reproduced based on high-resolution mapping data from cases of human gastric dysfunction (gastric re-entry, conduction blocks and ectopic pacemakers) in addition to normal test data. Each case was recorded and compared to the previously-presented simulated results. Qualitative and quantitative analyses were performed to define the accuracy showing \(\sim \) 1.8% difference, \(\sim \) 0.99 correlation, and \(\sim \) 0.04 normalised RMS error between experimental and simulated findings. These results reaffirm previous findings and these methods in unison therefore present a promising morphological-based methodology for advancing the understanding and clinical applications of EGG.     

Effect of signal length on the performance of independent component analysis when extracting the lambda wave

The aim of the study was to investigate the effect of signal length on the performance of a signal source separation method, independent component analysis (ICA), when extracting the visual evoked potential (EP) lambda wave from saccade-related electro-encephalogram (EEG) waveforms. A method was devised that enabled the effective length of the recorded EEG traces to be increased prior to processing by ICA. This involved abutting EEG traces from an appropriate number of successive trials (a trial was a set of waveforms recorded from 64 electrode locations in a study investigating saccade performance). ICA was applied to the saccade-related EEG and electro-oculogram (EOG) waveforms recorded from the electrode locations. One spatial and five temporal features of the lambda wave were monitored to assess the performance of ICA applied to both abutted and non-abutted waveforms. ICA applied to abutted trials managed to extract all six features across all seven subjects included in the study. This was not the case when ICA was applied to the non-abutted trials. It was quantitatively demonstrated that the process of abutting EEG waveforms was useful for ICA preprocessing when extracting lambda waves.     

Response of the electric activity in the human stomach to water and a solid meal

The gastric electric activity paces the contractions of the stomach and can be measured noninvasively using surface electrodes placed on the abdomen. In this paper, response of the gastric electric activity to water and a solid meal was investigated using surface electrodes. Applying an improved recording technique, power spectral analysis and statistical analysis methods, the gastric electric activity in preprandial, postwater and postprandial states was measured and analysed from ten healthy volunteers. To more convincingly show the effect of water and the solid meal on the gastric slow wave and gastric motility, simultaneous recordings of the cutaneous EGG and manometric activities in the stomach were performed in three gastroparetic patients. It was found that water does not induce contractions of the stomach, but the gastric slow wave amplitude increases (p<0·05) and frequency decreases (p<0·05). The solid meal results in an increase in both the amplitude (p<0·01) and the frequency (p<0·02) of the gastric slow wave. It appears that the postprandial EGG amplitude increases because of a combination of stomach contractions and gastric distension bringing the stomach closer to the recording electrodes. The slow wave frequency, however, correlates well with stomach contractions.     

On Non-Invasive Measurement of Gastric Motility from Finger Photoplethysmographic Signal

This article investigates the possibility of extracting gastric motility (GM) information from finger photoplethysmographic (PPG) signals non-invasively. Now-a-days measuring GM is a challenging task because of invasive and complicated clinical procedures involved. It is well-known that the PPG signal acquired from finger consists of information related to heart rate and respiratory rate. This thread is taken further and effort has been put here to find whether it is possible to extract GM information from finger PPG in an easier way and without discomfort to the patients. Finger PPG and GM (measured using Electrogastrogram, EGG) signals were acquired simultaneously at the rate of 100 Hz from eight healthy subjects for 30 min duration in fasting and postprandial states. In this study, we process the finger PPG signal and extract a slow wave that is analogous to actual EGG signal. To this end, we chose two advanced signal processing approaches: first, we perform discrete wavelet transform (DWT) to separate the different components, since PPG and EGG signals are non-stationary in nature. Second, in the frequency domain, we perform cross-spectral and coherence analysis using autoregressive (AR) spectral estimation method in order to compare the spectral details of recorded PPG and EGG signals. In DWT, a lower frequency oscillation (≈0.05 Hz) called slow wave was extracted from PPG signal which looks similar to the slow wave of GM in both shape and frequency in the range (0–0.1953) Hz. Comparison of these two slow wave signals was done by normalized cross-correlation technique. Cross-correlation values are found to be high (range 0.68–0.82, SD 0.12, R = 1.0 indicates exact agreement, p < 0.05) for all subjects and there is no significant difference in cross-correlation between fasting and postprandial states. The coherence analysis results demonstrate that a moderate coherence (range 0.5–0.7, SD 0.13, p < 0.05) exists between EGG and PPG signal in the “slow wave” frequency band, without any significant change in the level of coherence in postprandial state. These results indicate that finger PPG signal contains GM-related information. The findings are sufficiently encouraging to motivate further exploration of finger PPG as a non-invasive source of GM-related information.     

Filter Banks and Neural Network-based Feature Extraction and Automatic Classification of Electrogastrogram

Dysrhythmia in gastric myoelectrical activity has been frequently observed in patients with gastric motor disorders and gastrointestinal symptoms. The assessment of the regularity of gastric myoelectrical activity is of great clinical significance. The aim of this study was to develop an automated assessment method for the regularity of gastric myoelectrical activity from the surface electrogastrogram (EGG). The method proposed in this paper was based on the filter bank and neural network. First, the EGG signal was divided into frequency subbands using filter bank analysis. Second, a parameter called the subband energy ratio (SER) was computed for each subband signal. A multilayer perceptron neural network was then used to automatically classify the EGG signal into four categories: bradygastria, normal, tachygastria, and arrhythmia, using the SER as the input. The EGG recording was made using the standard method of electrogastrography by placing electrodes on the abdominal surface. The study was performed in 40 patients with various gastric motor disorders, ten healthy adults, and ten healthy children. The neural network was trained and tested using the EGG data obtained from the patients. The regularity of gastric myoelectrical activity was assessed based on the classification of the minute-by-minute EGG segments. Using the running spectral analysis method as a gold standard, the proposed automated method had an accuracy of 100% for the training set and 97% for the test set. It was concluded that the proposed method provides an accurate and automatic assessment of the regularity of gastric myoelectrical activity from the EGG. © 1999 Biomedical Engineering Society. PAC99: 8780-y, 8717-d, 0705Mh, 0270Hm     

胃动力信息检测系统的设计与实现

将胃电与胃机械活动相关的胃阻抗信息作为诊断依据,建立新的、更准确的胃动力评价方法.基于阻抗法设计了获取胃动力信息的系统,通过小波变换技术分离与胃电同步的胃蠕动信息.结果表明系统通过了调试,实现设计功能,采集到了阻抗和同步胃电信号.小波变换提取的信号排除了干扰,获得了胃蠕动信息.基于阻抗技术的方法可检测胃蠕动状态,联合同步胃电分析技术为临床检测提供了新方法.     

Physiological Response Patterns in Ulcer Patients: Phasic and Tonic Components of the Electrogastrogram

Physiological response patterns in duodenal ulcer patients, rheumatoid arthritics, and healthy subjects were investigated by recording electrogastrograms (EGGs) and several other physiological measures during presentations of mildly stressful stimuli. Ulcer patients were physiologically less reactive to the stimuli than the other groups; with the exception of increases in tonic galvanic skin potential (GSP) responses during arithmetic problems, ulcer patients were responsive only while viewing slides taken during autopsies when they evidenced increases in gastric activity and phasic GSP responses. The EGG proved to be a valuable measure for differentiating both the groups and the tasks. Both ulcer patients and healthy subjects demonstrated increases in phasic (frequency and amplitude) EGG activity during affective but not during cognitive stimuli, and ulcer patients and arthritics could be differentiated from healthy subjects on the basis of tonic (basal resting levels and displacement) EGG activity. These results were interpreted to surest the possibility that phasic and tonic components of the EGG may each be associated with different types of gastrointestinal activity.     

High-resolution Mapping of <Emphasis Type="Italic">In Vivo</Emphasis> Gastrointestinal Slow Wave Activity Using Flexible Printed Circuit Board Electrodes: Methodology and Validation

High-resolution, multi-electrode mapping is providing valuable new insights into the origin, propagation, and abnormalities of gastrointestinal (GI) slow wave activity. Construction of high-resolution mapping arrays has previously been a costly and time-consuming endeavor, and existing arrays are not well suited for human research as they cannot be reliably and repeatedly sterilized. The design and fabrication of a new flexible printed circuit board (PCB) multi-electrode array that is suitable for GI mapping is presented, together with its in vivo validation in a porcine model. A modified methodology for characterizing slow waves and forming spatiotemporal activation maps showing slow waves propagation is also demonstrated. The validation study found that flexible PCB electrode arrays are able to reliably record gastric slow wave activity with signal quality near that achieved by traditional epoxy resin-embedded silver electrode arrays. Flexible PCB electrode arrays provide a clinically viable alternative to previously published devices for the high-resolution mapping of GI slow wave activity. PCBs may be mass-produced at low cost, and are easily sterilized and potentially disposable, making them ideally suited to intra-operative human use.