Influence of recording instrumentation on the stimulus artifact tail in the surface acquisition of somatosensory evoked potentials.

Surface recorded somatosensory evoked potentials (SEPs) are neural signals elicited by an external stimulus. In the case of electrically induced SEPs, the artifact generated by the stimulation process can severely distort the signal. The artifact is characterized by a large impulse followed by a slowly decaying tail. In some cases, the artifact tail often lasts well into the initiation of the SEP making the determination of absolute latency very difficult. While the literature often states that the recording instrumentation plays a part in the generation of this artifact tail, no firm evidence has ever been presented. In this work, comparisons are made between three instrumentation systems (BJT, JFET and CMOS) with differing input impedances in an attempt to quantify the effects on the artifact tail. The conclusions from this investigation show that there is no significant interaction between the input impedance of the recording instrumentation and the duration of the artifact tail. Each amplifier type produced results with no significant statistical differences. It was also found that while stimulation amplitude has a weak effect on the artifact tail, the greatest contribution to variation has an inter-subject origin. Consequently, it is concluded that the time constant of the artifact tail must originate from other sources that are subject dependent.

Factors affecting the stimulus artifact tail in surface-recorded somatosensory-evoked potentials.

Surface-recorded somatosensory-evoked potentials (SEPs) are neural signals elicited by an external stimulus. In the case of electrically induced SEPs, the artifact generated by the stimulation process can severely distort the signal. In some cases, the artifact tail often lasts well into the initiation of the SEP making the determination of absolute latency very difficult. In this work, a new approach was taken to identify factors that affect the tail of the artifact. The methodology adopted was the development of a lumped electrical circuit model of the artifact generation process. While the modeling of the instrumentation hardware is relatively simple, this is not the case with tissue and electrode/skin interface effects. Consequently, this paper describes a novel tissue modeling approach that uses an autoregressive moving average (ARMA) parametric technique and an artificial neural network (ANN) to estimate tissue parameters from experimental data. This coupled with an estimation of the stimulation electrode-skin impedance completes the lumped circuit model. Simulink (The Mathworks Inc.) was used to evaluate the model under several different conditions. These results show that both the stimulation electrode-skin interface impedance and nature of the body tissue directly under the recording electrodes have a profound effect on the appearance of the stimulus artifact tail. This was verified by experimentally recorded data obtained from the median nerve using surface electrodes. Conclusions drawn from this work include that stimulation electrodes with low series capacitance should be used whenever possible to minimize the duration of the artifact tail.

Myo-electric signals to augment speech recognition.

It is proposed that myo-electric signals can be used to augment conventional speech-recognition systems to improve their performance under acoustically noisy conditions (e.g. in an aircraft cockpit). A preliminary study is performed to ascertain the presence of speech information within myo-electric signals from facial muscles. Five surface myo-electric signals are recorded during speech, using Ag-AgCl button electrodes embedded in a pilot oxygen mask. An acoustic channel is also recorded to enable segmentation of the recorded myo-electric signal. These segments are processed off-line, using a wavelet transform feature set, and classified with linear discriminant analysis. Two experiments are performed, using a ten-word vocabulary consisting of the numbers 'zero' to 'nine'. Five subjects are tested in the first experiment, where the vocabulary is not randomised. Subjects repeat each word continuously for 1 min; classification errors range from 0.0% to 6.1%. Two of the subjects perform the second experiment, saying words from the vocabulary randomly; classification errors are 2.7% and 10.4%. The results demonstrate that there is excellent potential for using surface myo-electric signals to enhance the performance of a conventional speech-recognition system.

Errors associated with the use of adaptive differential pulse code modulation in the compression of isometric and dynamic myo-electric signals.

Muscle activity produces an electrical signal termed the myo-electric signal (MES). The MES is a useful clinical tool, used in diagnostics and rehabilitation. This signal is typically stored in 2 bytes as 12-bit data, sampled at 3 kHz, resulting in a 6 kbyte s-1 storage requirement. Processing MES data requires large bit manipulations and heavy memory storage requirements. Adaptive differential pulse code modulation (ADPCM) is a popular and successful compression technique for speech. Its application to MES would reduce 12-bit data to a 4-bit representation, providing a 3:1 compression. As, in most practical applications, memory is organised in bytes, the realisable compression is 4:1, as pairs of data can be stored in a single byte. The performance of the ADPCM compression technique, using a real-time system at 1 kHz, 2 kHz and 4 kHz sampling rates, is evaluated. The data used include MES from both isometric and dynamic contractions. The percent residual difference (PRD) between an unprocessed and processed MES is used as a performance measure. Errors in computed parameters, such as median frequency and variance, which are used in clinical diagnostics, and waveform features employed in prosthetic control are also used to evaluate the system. The results of the study demonstrate that the ADPCM compression technique is an excellent solution for relieving the data storage requirements of MES both in isometric and dynamic situations.

A comparison of ECG cancellation techniques applied to the surface recording of somatosensory evoked potentials.

The use of somatosensory evoked potentials (SEPs) for determining spinal cord integrity in a clinical setting is well known. The poor signal-to-noise (SNR) ratio of such measurements has led to ensemble averaging being employed to extract the signal from the background noise. Over the thoracic region, the poor SNR is largely the result of interference generated by the cardiac musculature. Therefore, any reduction in the level of this cardiac interference will greatly improve the performance of any SEP monitoring system. Three methods were investigated as techniques to reduce the cardiac interference in SEP measurements. These were clipping, gating and adaptive noise cancellation (ANC). It was found that, although clipping and gating performed as well as, if not better than, ANC, these techniques were both very sensitive to the setting of a threshold level. The linear ANC scheme employed circumvented the problem associated with a threshold level, and with an order of 50, SNR improvement figures of approximately sixfold were achievable over the entire thoracic region.

Real-time compression of myoelectric data utilising adaptive differential pulse code modulation.

The myoelectric signal, obtained by either surface or needle electrodes, is used in many areas of clinical research and diagnosis. The conventional method of storing such information is in digitised form on a computer. However, the bandwidth of the signal and the required resolution result in large memory requirements. Adaptive differential pulse code modulation is investigated as a method of reducing the memory requirements for myoelectric data storage. In this scheme, a 12-bit sample is reduced to four bits, thus reducing the memory requirements by a factor of three. In reality, this compression ratio is closer to 4:1 owing to the fact that the widths of most memories are organised as multiples of eight bits.

Reduction of evoked potential measurement time by a TMS320 based adaptive matched filter.

In some clinical centres, somatosensory evoked potentials are used for the assessment of neurological function during surgical procedures on the spine. As these potentials are heavily contaminated in background noise, ensemble averaging coupled with invasive instrumentation is routinely used to enhance the signal. However, this procedure is very time consuming, often taking several minutes. In this paper, an adaptive matched filter has been used to dramatically reduce this measurement time to around 20 seconds, even when employing non-invasive surface electrodes. This filter has been implemented in real-time by using a TMS320C25 digital signal processor and results are presented for signal acquisition both at the L1-T12 and T5-T6 spinal levels. In addition, the adaptive nature of this filter allows the tracking of slowly changing parameters within the evoked potential with time.

Identification of noise sources in surface recording of spinal somatosensory evoked potentials.

The poor signal-to-noise ratio associated with the acquisition of evoked potentials is a well established fact. The problem is compounded if non-invasive techniques, using surface electrodes, are employed. The paper identifies several sources of noise associated with the acquisition of spinal somatosensory evoked potentials using surface electrodes. In addition, the relative contribution of these sources is determined experimentally for six spinal levels ranging from lower lumbar to upper thoracic. These data will prove useful in the design of digital signal processing schemes such as adaptive noise cancellation, where levels of uncorrelated noise severely limit system performance.

Modifications to hydra-gym equipment provide for clinically useful strength measurements.

Modifications were made to the Series III-311 Omnikinetic Hydra-Gym apparatus (Hydra-Fitness Industries Inc., Belton, TX) in an effort to provide relatively inexpensive yet clinically useful isokinetic strength measurements. Signal transmission and processing functions were installed to calculate relevant strength characteristics for knee extension/flexion motions. Results from the modified Hydra-Gym were compared with matched procedures on the Kin-Com (model 500-10, Chattecx Corporation, Chattanooga, TN). Subjects (N = 10) performed three maximal concentric right knee extension/flexion cycles at different resistance settings (2, 4, and 6) on the Hydra-Gym and the corresponding angular velocities on the Kin-Com (190, 125, and 40 degrees/sec, respectively). Each subject completed eight testing sessions, four on each dynamometer. A five-way analysis of variance on the peak torque data revealed no significant difference between dynamometers; however, there was a significant interaction between dynamometer and extension/flexion measurements. The variance in angle of peak torque data was significantly different between dynamometers, with Hydra-Gym showing superior reproducibility (p < 0.005). Mechanical differences between dynamometers produced slightly different, though comparable and consistent, measurement values. Care must be taken when evaluating hamstring/quadriceps ratios from the Hydra-Gym because the interaction noticed between dynamometer and extension/flexion may exaggerate muscle imbalances. In summary, modifications to the Hydra-Gym appear to provide clinicians with a reliable and clinically useful strength testing alternative.

Operator performance in myoelectric control of a multifunction prosthesis stimulator.

In this paper, a microcomputer-based system is described which facilitates the evaluation of different myoelectric control strategies. The system is based on an IBM-PC/AT microcomputer which generates and develops tracking targets, processes and displays the operator's response, and computes tracking performance. The dynamics of the prosthetic elbow, hand, and forearm are simulated in software which drives a stick figure in response to the operator's input. Performance is measured in terms of integral absolute error in target-response match. Control strategies to simulate different systems are easily modified through software and are evaluated under identical conditions. As a test of this evaluation system and to obtain some comparative data on control strategies, four strategies which are either in use or proposed for use are evaluated. The test procedure is divided into a control training session and a control evaluation session. Data collected from groups of three normally-limbed subjects per strategy over a period of five control evaluation sessions are presented. The results of this evaluation are discussed and improvements in the evaluation system suggested.