Gait control system for functional electrical stimulation using neural networks

In functional electrical stimulation (FES) systems for restoring walking in spinal cord injured (SCI) individuals, hand switches are the preferred method for controlling stimulation timing. Through practice the user becomes an ‘expert’ in determining when stimulation should be applied. Neural networks have been used to ‘clone’ this expertise but these applications have used small numbers of sensors, and their structure has used a binary output, giving rise to possible controller oscillations. It was proposed that a threelayer structure neural network with continuous function, using a larger number of sensors, including ‘virtual’ sensors, can be used to ‘clone’ this expertise to produce good controllers. Using a sensor set of ten force sensors and another of 13 ‘virtual’ kinematic sensors, a good FES control system was constructed using a three-layer neural network with five hidden nodes. The sensor set comprising three sensors showed the best performance. The accuracy of the optimum three-sensor set for the force sensors and the virtual kinematic sensors was 90% and 93%, respectively, compared with 81% and 77% for a heel switch. With 32 synchronised sensors, binary neural networks and continuous neural networks were constructed and compared. The networks using continuous function had significantly fewer oscillations. Continuous neural networks offer the ability to generate good FES controllers.     

Effects of functional electrical stimulation on the urethral closing muscles

The importance of the selection of patients with urinary incontinence suitable for treatment with functional electrical stimulation (f.e.s.) was tested. 40 patients with urinary incontinence of different kinds (mostly stress incontinence) were selected on the basis of urological, neurophysiological and urodynamic examinations. The positive response to optimal f.e.s. was a criterion for urodynamic selection. Mechanical properties of the unstimulated and electrically stimulated urethral wall tissue in the functional part of the urethra were studied. The differences between the acute effects of different types of stimulation, namely monophasic and biphasic f.e.s. with vaginal or different types of anal plugs and mechanical stimulation produced by these plugs, were stated. An attempt was made to quantify the success of treatment with f.e.s. Besides the routine urological, neurophysiological and urodynamic methods, the investigative methods of measuring the urethral pressure profile (u.p.p.) and measuring the pressure at one point of the urethra, by means of Mikro-Tip pressure transducers, were used in our study. Errors due to the dimensions of the measuring device causing a nonphysiological dilatation of the urethra were considered in the method of measuring the u.p.p.     

Injectable microstimulator for functional electrical stimulation

A family of digitally controlled devices is constructed for functional electrical stimulation in which each module is an hermetically sealed glass capsule that is small enough to be injected through the lumen of a hypodermic needle. The overall design and component characteristics of microstimulators that receive power and command signals by inductive coupling from a single, externally worn coil are described. Each device stores power between stimulus pulses by charging an electrolytic capacitor formed by its two electrodes, made of sintered, anodised tantalum and electrochemically activated iridium, respectively. Externally, a highly efficient class E amplifier provides power and digitally encoded command signals to control the amplitude, duration and timing of pulses from up to 256 such microstimulators.     

Systematic characterisation of silicon-embedded accelerometers for mechanomyography

Silicon soft suction sockets (roll-on sleeves) currently used in passive prostheses for below-elbow amputees could also be used in externally powered prostheses, enhancing their functionality and comfort. However, as it is extremely difficult to hold currently used electromyography (EMG) sensors in place reliably within a silicon socket, an alternative measurement of muscular activity as the control input is necessary. Mechanomyography (MMG) is the epidermal measurement of the low-frequency vibrations produced by a contracting muscle. MMG sensors do not have to be in direct contact with the skin. Moreover, the embedding of sensors in the roll-on sleeve may also solve attachment issues, making sensor placement flexible. Therefore the objective was to determine the feasibility of recording MMG signals using silicon-embedded, micro-machined accelerometers. Fifteen embedded accelerometers were excited with predefined vibration patterns. The signal-to-noise ratio (SNR) and frequency response of each sample were measured and compared with those of non-embedded accelerometers. The SNR of embedded samples (≊19 dB) was significantly higher than that of non-embedded samples (≊12 dB), owing to the considerable mechanical damping effect of the silicon in the 300–900 Hz bandwidth (p=0.0028). This has implications for the application of silicon-embedded accelerometers for externally powered prosthesis control.     

Error rate in five-state myoelectric control systems

A myoelectric control system is one in which the operation of a device is controlled by the electric potential produced by voluntary contraction of a muscle. The utilisation of a single muscle site for the 3-state control of a myoelectric prosthesis has proven beneficial for certain amputees. It has been proposed that a single muscle site could be used for 5-state control, perhaps for an elbow and hand. This paper presents a preliminary study of the error probability for 5-state control.     

Finite state control of functional electrical stimulation for the rehabilitation of gait

Finite state control is an established technique for the implementation of intention detection and activity co-ordination levels of hierarchical control in neural prostheses, and has been used for these purposes over the last thirty years. The first finite state controllers (FSC) in the functional electrical stimulation of gait were manually crafted systems, based on observations of the events occurring during the gait cycle. Subsequent systems used machine learning to automatically learn finite state control behaviour directly from human experts. Recently, fuzzy control has been utilised as an extension of finite state control, resulting in improved state detection over standard finite state control systems in some instances. Clinical experience over the last thirty years has been positive, and has shown finite state control to be an effective and intuitive method for the control of functional electrical stimulation (FES) in neural prostheses. However, while finite state controlled neural prostheses are of interest in the research community, they are not widely used outside of this setting. This is largely due to the cumbersome nature of many neural prostheses which utilise externally mounted gait sensors and FES electrodes. FES-based control of movement has been subject to the constraints of artificial sensor and FES actuator technologies. However, continued advances in natural sensors and implanted multi-channel stimulators are broadening the boundaries of artificial control of movement, driving an evolutionary process towards increasingly human-like control of FES-based gait rehabilitation systems.     

Quantitative weightbearing and gait evaluation of paraplegics using functional electrical stimulation

In this study functional electrical stimulation (FES) was used to generate supported gait in paraplegic patients with traumatic upper-motor-neuron lesions, between the T5 and T12 spinal levels. Four patients have so far been treated and studied over a period of one year. The quadriceps and gluteus maximus muscles are stimulated simultaneously to achieve a supported standing position, while hip, knee and ankle flexions are achieved, alternately for each leg by stimulating the shank surface at two selected locations (flexion reflex). The stimulus used has an intensity of 120V, duration of 0·3 ms and frequency of 24 Hz. The standing and walking behaviour of patients was monitored in the gait laboratory of the Loewenstein Hospital. The amount of weightbearing on each foot during standing was established by time integration, over a standard period of time, of the reaction forces, as measured by ‘Kistler’ force platforms, on which the patient was required to stand, while taking care to support his walking aids outside the platform area. The gait of the patients was evaluated by means of an electrical contact system in which time/distance parameters of the stride were measured. Some typical results indicate a walking velocity of 10 cm s⁻¹ and weightbearing on the feet during standing of 80 per cent of body weight. Computer processing of the data acquired was used to obtain objective evaluation of the patients' progress during their training period towards easier and more independent mobility.     

Control of a hand grasp neuroprosthesis using an electroencephalogram-triggered switch: demonstration of improvements in performance using wavepacket analysis

Volitionally modulated electroencephalographic (EEG) waves were monitored for the purpose of controlling a hand neuroprosthesis in people with tetraplegia. The region of the EEG signal spectrum monitored was the occipital alpha wave (8–13 Hz), and volitional modulation was achieved with the opening and closing of the eyes. In a set of 13 trials evaluated, a subject with tetraplegia successfully completed ten trials undertaking stimulated grasp and release using the EEG-triggered switch. EEG signal data recorded during the 13 trials were also post-processed off-line using wavepacket analysis. Following this signal processing, the speed and reliability of the EEG-triggered switch, when operated by the subject with tetraplegia, was significantly improved (p<0.002). Such improvements provide system performance that is likely to be acceptable to a neuroprosthesis user during activities of daily life.     

Open-loop position control of the knee joint using electrical stimulation of the quadriceps and hamstrings

The clinical acceptability of functional electrical stimulation (FES) as an aid for restoration of paraplegic gait is limited by the inability to accurately and repeatedly position the lower extremity. To gain insight into the causes of and possible solutions to this problem, the responses of the quadriceps and hamstrings to FES were studied in able-bodied subjects. Isometric torque was dependent on knee angle and changed unpredictably with time. An open-loop feedforward knee-joint position controller was also tested. The results demonstrated that it is beneficial to account for the dependence of torque on position, that modifications to this openloop controller might improve accuracy and that closed-loop control may be essential for functional restoration of gait.     

Modelling the response of scalp sensory receptors to transcranial electrical stimulation

Transcranial electrical stimulation of the brain cause considerable discomfort to the patient. The purpose of the study was to find out whether this could be affected by the choice of stimulation parameters. A spherical volume conductor model of the head and active compartmental models of a pyramidal motor nerve and scalp nociceptor were used in combination to simulate the scalp nociception to transcranial electrical stimulation. Scalp nociceptors were excited at distances of several centimetres from the electrodes. The size of the excited scalp area correlated with the length of the stimulation pulse. The area was 12.3, 20.4 and 26.0 cm², for a 10μs, 100μs and 1 ms constant current pulse, respectively. With a 100 μs constant current pulse, the threshold for motor excitation was 205 mA and, for nociception, it was 51 mA. There was no significant difference between constant current and capacitor discharge pulses or between electrodes of different sizes. The results imply that the use of very short stimulation pulses can reduce the pain. If a topical anaesthesia is used to reduce the pain, it has to be applied on a large area around the electrodes.     

Effects of some different pulse parameters on bladder inhibition and urethral closure during intravaginal electrical stimulation: an experimental study in the cat

Intravaginal electrical stimulation (IVS) was used in the cat to induce bladder inhibition and urethral closure. The purpose of the study was to compare the efficacy of alternating constant-voltage pulses of different durations (0·1, 0·2, 0·5 and 5 ms) with that of chopped alternating pulses, each consisting of a train of five 0·5 ms pulses with 0·5 ms pauses between. The voltage requirement for an equal biological effect was lowest for the 5 ms alternating pulse, whereas the pulse power dissipation was lowest for the 0·1 ms pulse, about 10 per cent of that for the 5 ms pulse. If both voltage requirement and power dissipation are taken into account, the 0·5 ms pulse was considered the most appropriate compromise. The chopped pulse was the least efficient stimulus, since the threshold voltage was comparable to that of the 0·5 ms duration alternating pulse, implying a five times higher power dissipation. For bladder inhibition, the optimal stimulation frequency for alternating pulses was 10 Hz, both in terms of threshold voltage and power consumption. For urethral closure the voltage requirement was lowest at 50 Hz but 20 Hz was preferable in terms of power dissipation for an equal, although submaximal, effect.     

Determining appropriate models for joint control using surface electrical stimulation of soleus in spinal cord injury

The mechanical impedance of the ankle joint during electrical stimulation of the soleus is studied by applying constant-velocity 10° angular perturbations to the ankle and measuring the resultant torque. Both neurologically intact subjects and spinal cord injured subjects are tested. Lumped, piecewise linear models are developed to predict the torque from the measured displacement and acceleration signals. The commonly used second-order mass-spring-dashpot model fails to predict the changes in torque that occur following imposed movements. A fiveelement, directionally-dependent piecewise linear model is much better at predicting the measured responses for velocities up to 50° s⁻¹. Numerical least squared error indentification techniques are used to estimate the model parameters for three neurologically intact and three spinal cord injured subjects. The average error between the model’s response and the measured response across all subjects is 10·9%. There is some evidence that a velocity-dependent non-linear model could produce better results than the directionally-dependent piecewise linear model.     

Short-term effects of functional electrical stimulation on motor-evoked potentials in ankle flexor and extensor muscles

Stimulating sensory afferents can increase corticospinal excitability. Intensive use of a particular part of the body can also induce reorganization of neural circuits (use-dependent plasticity) in the central nervous system (CNS). What happens in the CNS when the nerve stimulation is applied in concert with the use of particular muscle groups? The purpose of this study was to investigate short-term effects of electrical stimulation of the common peroneal (CP) nerve during walking on motor-evoked potentials (MEPs) in the ankle flexors and extensors in healthy subjects. Since the stimulation was applied during the swing phase of the step cycle when the ankle flexors are active, this is referred to as functional electrical stimulation (FES). The following questions were addressed: (1) can FES during walking increase corticospinal excitability more effectively than passively received repetitive nerve stimulation and (2) does walking itself improve the descending connection. FES was delivered using a foot drop stimulator that activates ankle dorsiflexors during the swing phase of the step cycle. MEPs in the tibialis anterior (TA) and soleus muscles were measured before, between, and after periods of walking with or without FES, using transcranial magnetic stimulation. After 30 min of walking with FES, the half-maximum peak-to-peak MEP (MEP_h) in the TA increased in amplitude and this facilitatory effect lasted for at least 30 min. In contrast, walking had no effects on the TA MEP_h without FES. The increase in the TA MEP_h with FES (~40%) was similar to that with repetitive CP nerve stimulation at rest. The soleus MEP_h was also increased after walking with FES, but not without FES, which differs from the previous observation with CP nerve stimulation at rest. With FES, the TA silent period at MEP_h was unchanged or slightly decreased, while it increased after walking without FES. Increased cortical excitability accompanied by unchanged cortical inhibition (no changes in the silent period with FES) suggests that FES did not simply increase general excitability of the cortex, but had specific effects on particular cortical neurons.     

Changes in expired air composition during electrical stimulation of the ventral medulla

Department of Physiology of Visceral Systems, Institute of Experimental Medicine, Academy of Medical Sciences, Saint Petersburg. Translated from Byulleten' Éksperimental'noi Biologii i Meditsiny, Vol. 113, No. 4, pp. 339–340, April, 1992.     

呼吸电刺激训练改善膈肌和腹肌功能治疗功能性便秘的作用

目的　观察呼吸电刺激训练对功能性便秘患者的治疗作用。 方法　纳入功能性便秘患者共46例,按随机数字表法分为治疗A组、治疗B组和对照组。A组采用呼吸电刺激和盆底生物反馈治疗联合口服乳果糖治疗,共15例;B组进行盆底生物反馈联合口服乳果糖治疗,共16例;对照组只口服乳果糖治疗,共15例。治疗前和治疗2周后,评估功能性便秘严重程度,并评估膈肌和腹肌增厚率、膈肌活动度。 结果　治疗2周后,各组患者便秘严重程度明显下降（P<0.05）,A组膈肌和腹肌增厚率、膈肌活动度较治疗前改善（P<0.05）;与B组及对照组比较,A组便秘严重程度评分明显改善（P<0.05）,膈肌和腹肌增厚率、膈肌活动度改善程度改善明显（P<0.05）。 结论　呼吸电刺激训练可增强盆底生物反馈治疗改善功能性便秘的治疗效果。     

Microprocessor-control of drug infusion for automatic blood-pressure control

A microprocessor-controlled infusion pump is described for the automatic control of hypotension or antihypertensive treatment with sodium nitroprussid. The system requires the continuous monitoring of blood pressure as an input signal to the microprocessor, the latter regulating blood pressure to the desired level by stepwise changes in infusion rate. The infusion pump proved valuable for intensive care and neurosurgery under controlled hypotension.     

The effect of previous weight training and concurrent weight training on endurance for functional electrical stimulation cycle ergometry

Forty-five paraplegic subjects participated in three series of experiments to examine the interrelationships between previous weight training, concurrent weight training and muscle strength and endurance during cycle ergometry elicited by functional electrical stimulation (FES). When subjects only underwent isokinetic weight training (series 1) three times per week on the quadriceps, hamstring and gluteus maximus groups for 12 weeks, strength increased linearly with time for all three muscle groups from an initial average of 17 N to 269 N at the end of training, a 15-fold increase. In the second series of experiments, different groups of subjects either underwent no strength training prior to cycle ergometry or underwent strength training of these same three muscle groups for 2 weeks, four weeks, or 6 weeks prior to cycle ergometry. Any strength training was effective in increasing endurance for cycle ergometry. However, the rate of increase in endurance during cycle ergometry with no prior strength training was only 5 min per week, whereas the rate of increase in cycle endurance during ergometry was 14.6, 25.0, and 33.3 min per week increase in endurance after strength training for 2.4 and 6 weeks, respectively. When weight training was accomplished during FES cycle ergometry (concurrently) in a third series of experiments, there was an additional increase in endurance during cycling if strength training was concurrently accomplished. With no weight training, endurance increased 23 min per week, whereas with concurrent weight training at three times per week, endurance increased during cycling by 41.6 min per week. The results of these experiments seem to show a clear advantage of weight training concurrently and before FES cycle ergometry. Results are given as mean (SD).     

Standing-up exerciser based on functional electrical stimulation and body weight relief

The goal of the present work was to develop and test an innovative system for the training of paraplegic patients when they are standing up. The system consisted of a computer-controlled stimulator, surface electrodes for quadricep muscle stimulation, two knee angle sensors, a digital proportional-integrative-derivative (PID) controller and a mechanical device to support, partially, the body weight (weight reliever (WR)). A biomechanical model of the combined WR and patient was developed to find an optimum reference trajectory for the PID controller. The system was tested on three paraplegic patients and was shown to be reliable and safe. One patient completed a 30-session training period. Initially he was able to stand up only with 62% body weight relief, whereas, after the training period, he performed a series of 30 standing-up/sitting-down cycles with 45% body weight relief. The closed-loop controller was able to keep the patient standing upright with minimum stimulation current, to compensate automatically for muscle fatigue and to smooth the sitting-down movement. The limitations of the controller in connection with a highly nonlinear system are considered.     

Effect of tactile stimulation pulse characteristics on sensation threshold and power consumption

The psychophysical responses of human subjects to vibratory tactile stimulation of the skin were investigated experimentally. The parameters, of the waveform important to the minimization of power consumed by the tactile array of electromechanical vibrators and the maximization of the skin sensitivity to the stimulus were explored to develop optimum stimulation. Parameters investigated included the amplitude, frequency, and duty cycle of the current waveform used to drive the vibrators as well as the number of pulses per stimulating burst and the recovery time between bursts. Graphical techniques were used to determine, the optimal combination of the parameters which gave a stimulus that excited the skin to above tactile threshold while maintaining at a relative minimum the power required for the stimulus. The optimal stimulation waveform contains a burst of 10 rectangular pulses of 4% duty cycle separated by a period of nonstimulation of 2 s. Such a waveform can elicit a sensitivity of 29.4 mA⁻¹ consuming only 55 μW of power.