Reliability of neural-network functional electrical stimulation gait-control system

Functional electrical stimulation (FES) has been used for restoring walking in spinal-cord injured (SCI) persons. Using artificial intelligence (AI), FES controllers have been developed that allow the automatic phasing of stimulation, to replace the function of hand or heel switches. However, there has been no study to evaluate the reliability of these AI systems. Neural networks were used to construct FES controllers to control the timing of stimulation. Different numbers of sensors in the sensor set and different numbers of data points from each sensor were used. Two incomplete-SCI subjects were recruited, and each was tested on three separate occasions. The results show the neural-network controllers can maintain a high accuracy (around 90% for the two- and three-sensor groups and 80% for the one-sensor group) over a period of six months. Two or three sensors were sufficient to provide enough information to construct a reliable FES control system, and the number of data points did not have any effect on the reliability of the system.     

Reliability of neural-network functional electrical stimulation gait-control system

Functional electrical stimulation (FES) has been used for restoring walking in spinal-cord injured (SCI) persons. Using artificial intelligence (Al), FES controllers have been developed that allow the automatic phasing of stimulation, to replace the function of hand or heel switches. However, there has been no study to evaluate the reliability of these Al systems. Neural networks were used to construct FES controllers to control the timing of stimulation. Different numbers of sensors in the sensor set and different numbers of data points from each sensor were used. Two incomplete-SCI subjects were recruited, and each was tested on three separate occasions. The results show the neural-network controllers can maintain a high accuracy (around 90% for the two- and three-sensor groups and 80% for the onesensor group) over a period of six months. Two or three sensors were sufficient to provide enough information to construct a reliable FES control system, and the number of data points did not have any effect on the reliability of the system.     

Damage in peripheral nerve from continuous electrical stimulation: Comparison of two stimulus waveforms

The propensity for two types of charge-balanced stimulus waveforms to induce injury during eight hours of continuous electrical stimulation of the cat sciatic nerve was investigated. One waveform was a biphasic, controlled-current pulse pair, each phase 50 μs in duration, with no delay between the phases (‘short pulse’, selected to excite primarily large axons), whereas in the second type each phase was 100 μs in duration, with a 400 μs delay between the phases (selected to excite axons of a broader spectrum of diameters). The sciatic nerve was examined for early axonal degeneration (EAD) seven days after the session of continuous stimulation. With both waveforms, the threshold stimulus current for axonal injury was greater than the current required to excite all of the nerve's large axons. The correlation between simple stimulus parameters and the amount of EAD was poor, especially with the ‘short pulse’ waveform, probably due to variability between animals. When the stimulus was normalised with respect to the current required to fully recruit the large axons, a good association between damage and stimulus amplitude emerged. The damage threshold was higher for the ‘short pulse’ waveform. The implications for clinical protocols are discussed.     

Finger interaction during maximal radial and ulnar deviation efforts: experimental data and linear neural network modeling

The purpose of this study was to characterize finger interactions during radial/ulnar deviation, including interactions with flexion movements. Subjects performed single-finger and multi-finger maximal voluntary contraction (MVC), and maximal forces and various indices of interaction among the fingers were quantified. MVCs in radial/ulnar deviation were 50–80% as strong as in flexion. Along with the ‘master’ fingers (i.e., those explicitly instructed to produce force), substantial force production was also observed in ‘slave’ fingers (i.e., those not explicitly instructed to produce force), a phenomenon termed: force ‘enslaving’. In addition, a drop in MVC during multi-finger tasks as compared to single finger tasks (force ‘deficit’) was also observed. A previously unreported phenomenon that we term: ‘preferred direction enslaving’ was also apparent; both master and slave fingers produced force in the instructed direction with a non-zero perpendicular component. Due to the architectural separation of the involved muscles, preferred direction enslaving provides strong evidence that enslaving results from neural rather than biomechanical factors. A final new phenomenon: ‘negative deficit’, or force ‘facilitation’ was observed in 46.4% of the trials in 21 out of 23 subjects during multi-finger lateral efforts and was further demonstrative of extensive interconnection among neurons serving hand muscles. The data were modeled with high accuracy (∼4% mean square error) using a linear neural network with motor ‘commands’ as inputs and finger forces as outputs. The proposed network, equivalent to linear regression, can be used to determine the extent to which finger forces are influenced by peripheral constraints during functional prehensile activities.     

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.     

Silicon neuron simulation with SPICE: tool for neurobiology and neural networks

The paper deals with computer simulations of ‘silicon neurons’, which are assemblies of CMOS circuits that generate the equivalents of the ionic currents and of the action potentials of real (biological) neurons. The circuit simulation program SPICE is used to simulate the generation of action potentials by a silicon neuron. Moreover, the equivalent circuits of silicon synapses are described and the behaviours of simple two- and three-neuron networks are analysed. Implications for the areas of neurobiology and formal neural networks are briefly considered.     

Practice-related improvements in posture control differ between young and older adults exposed to continuous, variable amplitude oscillations of the support surface

Healthy older adults were repeatedly exposed to continuous, variable amplitude oscillations of the support surface to determine (1) whether age affects the capacity for postural motor learning under continuous perturbation conditions with limited predictability and (2) whether practice leads to modifications in the control strategy used to maintain balance in older adults. During training, a translating platform underwent 45-s trials of constant frequency (0.5 Hz) and seemingly random amplitude oscillations (range ±2 to 15 cm). The middle 15 s of each trial contained the same sequence of oscillation amplitudes. This repeated middle segment was the same as the repeated segment used in Van Ooteghem et al. (Exp Brain Res 187(4): 603–611, 2008) and was therefore used for analyses. To examine learning, participants performed a retention test following a 24-h delay. Kinematic data were used to derive spatial and temporal measures of whole body centre of mass (COM), trunk, thigh, and shank segment orientation, and ankle and knee angle from performance during the repeated middle segment. Results showed that with training, older adults maintained the capacity to learn adaptive postural responses in the form of improved temporal control of the COM and minimization of trunk instability at a rate comparable to young adults. With practice, however, older adults maintained a more rigid, ‘platform-fixed’ control strategy which differed from young adults who shifted towards ‘gravity-fixed’ control and decreased COM motion. This study provides important insight into the ability of older adults to demonstrate longer-term improvements in postural regulation.     

Non-constraining sleep/wake monitoring system using bed actigraphy

This paper introduces a new method, bed actigraphy (BACT) for user-friendly sleep-wake monitoring. BACT provides a non-intrusive acquisition of activity data, and in particular does not require that sensors be attached to the subject’s body. The system consists of four load-sensing cells supporting the bed, an A/D converter, and a microcontroller with appropriate software. The performance of BACT was compared to that of standard polysomnography (PSG) recordings and wrist-worn actigraphy (ACT). Ten normal volunteers underwent overnight PSG recordings and were examined simultaneously with BACT and ACT. An automatic scoring algorithm scored each 30-s epoch of the BACT recordings for either ‘Wake’ or ‘Sleep.’ A sleep specialist manually scored the PSG recordings, and the results were divided into ‘Wake’ and ‘Sleep’ categories. The three methods showed a significant correlation when compared with in the contingency test. The mean epoch-by-epoch agreements between the BACT and PSG, ACT and PSG, and BACT and ACT recordings were 95.2, 92.9, and 94.3%, respectively. The mean absolute differences in sleep percentage (SP) between them were 1.8 ± 0.82, 3.4 ± 1.45, and 1.9 ± 1.16 %, respectively. BACT differentiation of the ‘Wake’ and ‘Sleep’ stages proved to be sufficiently robust, and its results were comparable to PSG analysis. This finding supports the experimental and clinical value of bed-activity monitoring during sleep.     

Command control for functional electrical stimulation hand grasp systems using miniature accelerometers and gyroscopes

Recent commercially available miniature sensors have the potential to improve the functions of functional electrical stimulation (FES) systems in terms of control, reliability and robustness. A new control approach using a miniature gyroscope and an accelerometer was studied. These sensors were used to detect the linear acceleration and angular velocity of residual voluntary movements on upper limbs and were small and easy to put on. Five healthy subjects and three cervical spinal cord injured subjects were recruited to evaluate this controller. Sensors were placed on four locations: the shoulder, upper arm, wrist and hand. A quick forward-and-backward movement was employed to produce a distinctive waveform that was different from general movements. A detection algorithm was developed to generate a command signal by identifying this distinctive waveform through the detection of peaks and valleys in the sensor's signals. This command signal was used to control different FES hand grasp patterns. With a specificity of 0.9, the sensors had a success rate of 85–100% on healthy subjects and 82–97% on spinal cord injured subjects. In terms of sensor placement, the gyroscope was better as a control source than the accelerometer for wrist and hand positions, but the reverse was true for the shoulder.     

Control of hand shaping in response to object shape perturbation

This study assessed how hand shaping responds to a perturbation of object shape. In blocked trials (80% of total), subjects were instructed to reach, to grasp and lift a concave or a convex object. In perturbed trials (20% of total), a rotating device allowed for the rapid change from the concave to the convex object or vice versa. In this situation subjects grasped the last presented object. Flexion/extension at the metacarpal-phalangeal and proximal interphalangeal joints of all digits was measured by resistive sensors embedded in a glove. In the blocked condition we found that most joints of the fingers were modulated by the type of the to-be-grasped object during the reach. When object shape was perturbed, reach duration was longer and angular excursion of all fingers differed with respect to blocked trials. For the ‘convex → concave’ perturbation, a greater degree of finger extension was found than during the blocked ‘concave’ trials. In contrast, for the ‘concave → convex’ perturbation, fingers were more flexed than for the blocked ‘convex’ trials. The thumb reacted to the perturbation showing a similar pattern (i.e., over-flexion with respect to the blocked trials) regardless the ‘direction’ of the perturbation. The present results suggest that applying an object shape perturbation during a reach-to-grasp action determines a reorganization of all digits. This pattern is suggestive of a control strategy, which assigns to opposing digits different roles.     

Regulating knee joint position by combining electrical stimulation with a controllable friction brake

Hybrid FES gait restoration systems which combine stimulation with controllable mechanical damping elements at the joints show promise for providing good control of limb motion despite variations in muscle properties. In this paper we compared three controllers for position tracking of the free swinging shank in able-bodied subjects. The controllers were open-loop (OL), proportional-derivative closed-loop (PD), and bang-bang plus controlled-brake control (CB). Both OL and PD controllers contained a forward path element, which inverted a model of the electrically stimulated muscle and limb system. The CB control was achieved by maximally activating the appropriate muscle group and controlling the brake to be a “moving-wall” against which the limb pushed. The CB control resulted in superior tracking performance for a wide range of position tracking tasks and muscle fatigue states but required no calibration or knowledge of muscle properties. The disadvantages of CB control include excess mechanical power dissipation in the brake and impact forces applied to the skeletal system.     

Electrotactile stimuli delivered across fingertips inducing the Cutaneous Rabbit Effect

Previous studies have been unable to induce the Cutaneous Rabbit Effect (CRE) when the most likely perceived location of the illusory stimulus is on a non-continuous skin area. To determine whether the CRE could be elicited when each of the delivered stimuli were on non-continuous skin areas, we developed a new electrotactile stimulation paradigm attempting to induce the CRE across the fingertips. Though our stimulation paradigm differed from classic reduced CRE paradigms through the use of electrotactile stimuli, focusing the subject attention to a ‘likely’ illusory site, and the inclusion of a fourth stimulation site (two stimuli after the illusory stimulus), these factors were not the cause of the illusory effect we observed. Experiments conducted on the forearm validated that our paradigm elicited similar results to those reported in previous CRE studies that used either 3-stimulation-point mechanical or electrotactile stimuli with subject attention focused on the ‘likely’ illusory site. Across the fingertips, we observed an increase in stimulus mislocalization onto the middle fingertip, the ‘likely’ perceived location of the illusory stimuli, under Illusory Rabbit Trains compared to the Motion Bias Trains. Because the Motion Bias Trains should not induce a perceived location shift of the illusory stimulus but stimulates the adjacent digits in a similar way to the Illusory Rabbit Trains, differences observed between their mislocalization rates between these trains indicate that the CRE can be induced across the fingertips. These results provide the first evidence that the CRE can ‘jump’ when the stimuli occur across non-continuous skin areas.     

Neuromuscular stimulation selectivity of multiple-contact nerve cuff electrode arrays

The feasibility of an electrical stimulation method selectively for activating skeletal muscles innervated by a common peripheral nerve trunk has been investigated. The method utilises ‘snugly’ fitting nerve cuffs that incorporate an array of 12 electrodes. These electrodes have been tested as four longitudinally aligned tripoles (located 90° apart on the cuff inner surface). In acute experiments on rabbit sciatic nerves, we have found that tripolar stimulation with this implant system is in general highly selective. ‘Field steering’, wherein a subthreshold transverse current is used in combination with a longitudinal tripolar current, tends to increase the selectivity of stimulation. On a rabbit sciatic nerve, a combination of adjacent longitudinal tripoles of the 12 electrode array generally yields a stimulation performance similar to that which would be expected if a 24 electrode array is used. This system may find a use in functional neuromuscular stimulation applications which require highly selective control over multiple muscles.     

Intrinsic joint kinematic planning. I: Reassessing the Listing’s law constraint in the control of three-dimensional arm movements

This study tested the validity of the assumption that intrinsic kinematic constraints, such as Listing’s law, can account for the geometric features of three-dimensional arm movements. In principle, if the arm joints follow a Listing’s constraint, the hand paths may be predicted. Four individuals performed ‘extended arm’, ‘radial’, ‘frontal plane’, and ‘random mixed’ movements to visual targets to test Listing’s law assumption. Three-dimensional rotation vectors of the upper arm and forearm were calculated from three-dimensional marker data. Data fitting techniques were used to test Donders’ and Listing’s laws. The coefficient values obtained from fitting rotation vectors to the surfaces described by a second-order equation were analyzed. The results showed that the coefficients that represent curvature and twist of the surfaces were often not significantly different from zero, particularly not during randomly mixed and extended arm movements. These coefficients for forearm rotations were larger compared to those for the upper arm segment rotations. The mean thickness of the rotation surfaces ranged between ≈1.7° and 4.7° for the rotation vectors of the upper arm segment and ≈2.6° and 7.5° for those of the forearm. During frontal plane movements, forearm rotations showed large twist scores while upper arm segment rotations showed large curvatures, although the thickness of the surfaces remained low. The curvatures, but not the thicknesses of the surfaces, were larger for large versus small amplitude radial movements. In conclusion, when examining the surfaces obtained for the different movement types, the rotation vectors may lie within manifolds that are anywhere between curved or twisted manifolds. However, a two-dimensional thick surface may roughly represent a global arm constraint. Our findings suggest that Listing’s law is implemented for some types of arm movement, such as pointing to targets with the extended arm and during radial reaching movements.     

Development of visual control in stepping down

Stepping down at a change of height is a fundamental part of human locomotion. At a novel step, this requires the transformation of visual information about a depth change into a stepping movement of appropriate size. However, little is known about this process or its development. We studied adults, 3- and 4-year-old children stepping down a single stair of variable height. We assessed how well stepping down was scaled to stair height using several kinematic measures. Of these, ‘kneedrop’ and ‘toedrop’ describe how far the leg has descended by the time it begins to ‘swing in’ in preparation for landing; and ‘toeheight (speedpeak)’ describes where the toe begins to decelerate. If visually controlled, their values should scale to the height of the stair. Under normal visual conditions, children scaled these movements to stair height as well as adults. In a second condition, participants closed their eyes just before stepping down to remove visual feedback during the step. Adults’ steps were barely affected. For 4-year olds, only toeheight (speedpeak) decreased. For 3-year olds, both toedrop and toeheight (speedpeak) scaled less well to stair height than normal. The results suggest that visuomotor processes for fine-tuned stepping control develop remarkably early, but are initially dependent on visual feedback.     

Texture analysis of technegas lung ventilation images

Technegas lung ventilation images sometimes have ‘hot spots’, particularly in patients with respiratory disease. A novel technique is presented for quantifying this ‘spottiness’ using morphological texture analysis. A set of 32 images from patients with various respiratory diseases is studied. Images are filtered at a range of scales using morphological opening, and the slopes of image metrics versus structuring element size are used as texture parameters. The results are compared with the opinions of three experienced nuclear medicine physicians who have classified the images into two groups, ‘spotty’ and ‘non-spotty’, and have ranked the former. For the spotty images, the computer and observer ranks are compared; the highest correlation is r_s=0.66 (p=0.01) for a single parameter, andr _s =0.71 (p<0.01) for a combination of two parameters. Using a pair of parameters, 83% and 90% correct classification rates are obtained for the spotty and non-spotty classes, respectively. It is concluded that these texture parameters provide a useful measure of image spottiness, and it is demonstrated that this technique is superior to previously published methods. The practical value of the technique is illustrated using two applications.     

Analysis of Electrode Configurations for Measuring Cardiac Tissue Conductivities and Fibre Rotation

This paper describes a multi-electrode grid, which could be used to determine cardiac tissue parameters by direct measurement. A two pass process is used, where potential measurements are made, during the plateau phase of the action potential, on a subset of these electrodes and these measurements are used to determine the bidomain conductivities. In the first pass, the potential measurements are made on a set of ‘closely-spaced’ electrodes and the parameters are fitted to the potential measurements in an iterative process using a bidomain model and a solver based on a modified Shor's r-algorithm. This first pass yields the extracellular conductivities. The second pass is similar except that a ‘widely-spaced’ electrode set is used and this time the intracellular conductivities are recovered. In addition, it is possible to determine the fibre rotation throughout the tissue, since the bidomain model used here is able to include the effects of fibre rotation.In the simulation studies presented here, the model is solved with known conductivities, on each of the two subsets of electrodes, to generate two sets of ‘measured potentials.’ Conductivities are then recovered by solving an inverse problem based on the measured potentials, to which various levels of noise are added. For example, simulations in the first pass are performed using an electrode spacing of 500 μm, for a situation where the longitudinal and transverse space constants are 769 and 308 μm, respectively. These give very accurate average percentage relative errors for the longitudinal and transverse extracellular conductivities, over five simulations with 1% noise added, of 0.3 and 0.2%. Twenty-five second pass simulations, on a 1 mm grid, yield average percentage relative errors of 3.8, 2.6 and 1.4% for the corresponding intracellular values and the fibre rotation angle, respectively.     

Bioenergetics and biomechanics of cycling: the role of ‘internal work’

The ‘dissection’ of energy expenditure of cycling into the metabolic equivalent of the different forms of mechanical work done, inaugurated 30 years ago by di Prampero and collaborators, has been much debated in the last few decades. The mechanical internal work, particularly, which is currently associated to the movement of the lower limbs, has been approached, estimated and discussed in several different ways and there is no agreed consensus on its role in cycling. This paper, through re-processing previously published data of oxygen consumption during pedalling at different frequency, external load and limb mass, proposes a model equation and a multiple non-linear regression as the method to assess the internal work of cycling. With that tool a very consistent metabolic equivalent of the internal work is obtained. However, a software simulation of pedalling limbs showed, as suggested in the literature, that the link with the chain ring allows the system to passively revolve forever, after an initial push. This result challenges the very existence of the ‘kinematic internal work’ of cycling. We conclude and suggest that the ‘viscous internal work’, an often neglected and almost unmeasurable portion of the internal work that could be proportional to the ‘kinematic’ form, is responsible for the extra metabolic expenditure as measured when the pedalling frequency of cycling increases.     

Age Differences in Genetic and Environmental Variations in Stress-Coping During Adulthood: A Study of Female Twins

The way people cope with stressors of day to day living has an important influence on health. The aim of the present study was to explore whether genetic and environmental variations in stress-coping differ over time during adulthood. The brief COPE was mailed to a large sample of the UK female twins (N = 4,736) having a wide range of age (20–87 years). Factor analyses of the items of the brief COPE yielded three coping scales: ‘Problem-Solving’, ‘Support Seeking’, and ‘Avoidance’. Monozygotic and dizygotic twin correlations tended to become lower with age for all three scales, suggesting that unique environmental factors may become more important with age during adulthood. Model-fitting results showed that relative influences of unique environmental factors increased from 60 % at age 20 years to 74% at age 87 years for ‘Problem-Solving’ and 56 % at age 20 years to 76% at age 87 years for ‘Avoidance’. During the same age period, genetic factors decreased from 40 to 26 % for ‘Problem-Solving’ and from 44 to 24 % for ‘Avoidance’. For ‘Seeking Support’, the magnitude of genetic and unique environmental factors was not significantly different across the adulthood. For all three scales, shared environmental effects were negligible. Overall, our findings implicate that the effects of environment that stem from idiosyncratic experience of stressful life events accumulate and become increasingly important in adulthood.     

Reasons women give for abortion: a review of the literature

The aim was to identify from empirical research that used quantitative or qualitative methods the reasons women give for having an abortion. A search was conducted of peer-reviewed, English language publications indexed in eight computerized databases with publication date 1996–2008, using keywords ‘abortion’ and ‘reason’ (Medline: ‘induced abortion’ OR ‘termination of pregnancy’ OR ‘elective abortion’ and ‘reason’). Inclusion criteria were empirical research on humans that identified women’s reasons for undergoing an abortion, conducted in ‘high-income’ countries. 19 eligible papers were found. Despite variation in methods of generating, collecting, and analysing reasons, and the inadequacy of methodological detail in some papers, all contributed to a consistent picture of the reasons women give for having an abortion, with three main categories (‘Woman-focused’, ‘Other-focused’, and ‘Material’) identified. Ambivalence was often evident in women’s awareness of reasons for continuing the pregnancy, but abortion was chosen because continuing with the pregnancy was assessed as having adverse effects on the life of the woman and significant others. Women’s reasons were complex and contingent, taking into account their own needs, a sense of responsibility to existing children and the potential child, and the contribution of significant others, including the genetic father.