10:30
Motor Control, Neuro Control & Patient Models III
10:30
15 mins
|
EMG SPECTRAL PEAKS RECORDED DURING VIBRATION EXERCISE
Lin Xu, Chiara Rabotti, Massimo Mischi
Abstract: Resistance training (RT) is a widely accepted training option for improving neuromuscular performance. However, the high levels of load required in RT limit its applicability in rehabilitation programs. Vibration exercise (VE) is a relatively new and promising training modality, providing several important advantages over conventional strength training. To investigate the underlying mechanism elicited by VE, surface electromyography (EMG) has been widely used to measure the level of neuromuscular activity during VE. However, the EMG spectrum recorded during VE shows sharp peaks at the vibration frequency, whose interpretation (motion artifacts or muscle activity) remains controversial. The aim of the present study is to clarify the nature of the spectral peaks observed during VE. To this end, the conduction velocity of the vibration frequency component (CVVF) was estimated and compared with the CV of the entire EMG (CVEMG) and the acceleration signal (CVACC).
|
10:45
15 mins
|
APPLICATIONS OF VIDEO THERMOGRAPHY AND LASER DOPPLER IMAGING TO INVESTIGATE SMALL FIBER NEUROPATHY
Yusang Wu, Mariska Nieuwenhoff, Frank Huygen, Frans van der Helm, Sjoerd Niehof, Alfred Schouten
Abstract: Small fiber neuropathy (SFN) is a peripheral nerve disease that preferentially or selectively affects small nerve fibers and their functions. Diabetes is known as one of the major causes of SFN. Patients with SFN suffer from a combination of symptoms, including pain, numbness and vascular dysfunction. Early diagnosis of SFN and thus early treatment are crucial to prevent the development of SFN, and to reduce medial costs and usage of healthcare resources. Today, however, no good test or tool is available to identify SFN in an early stage.
Recent studies reported that thermally sensitive small fibers in the skin were important in the vasodilation response to local skin warming [1]. Our project aims to develop a novel method using non-contact and non-invasive technique to assess small nerve fiber function and as such to realize quantitative diagnosis of SFN in an early stage. In this study, the application of the method was demonstrated in healthy subjects.
Ten subjects participated in the study (20-30 years old). The dorsal side of the subjects’ hands was heated to 42 ℃ at different rising rates with a medical infrared lamp. The thermal response of the skin was evaluated based on two signals: (1) the skin temperature, measured with a video thermography camera and (2) the skin blood perfusion, measured with a laser Doppler imager. To demonstrate the control mechanism of small fibers, the skin temperature and the skin blood perfusion were investigated, and a control model was built using system identification technique to simulate the skin response.
Our study provided a comprehensive description of the skin thermal response on the dorsal side of hands. Next studies will be carried out in patients with SFN who are supposed to have a delayed/attenuated response to thermal stimuli.
Reference
[1] J. Johnson and D. Kellogg, “Local thermal control of the human cutaneous circulation”, Journal of Applied Physiology, Vol. 109, pp. 1229-1238, 2010
|
11:00
15 mins
|
SHORT AND LONG LATENCY REFLEXES DURING TRANSIENT AND CONTINUOUS PERTURBATIONS: WHAT IS WHAT?
Alfred Schouten, Ivo Ulrich, Winfred Mugge
Abstract: In many everyday activities such as steering a bicycle, unpredictable mechanical disturbances trigger quick involuntary and situation-specific muscle responses, i.e. reflexes. Research investigating upper limb reflexes via transient perturbations elicit muscle activations typically observed as two distinct electromyographic (EMG) responses: a stereotyped and task-independent short-latency M1 and a task-dependent long-latency M2. In contrast, studies using continuous perturbations and system identification do suggest task dependency of short-latency pathways. The brief transient position perturbations allow for straightforward reflex assessment, but deprive the subject of natural interaction with the manipulator as the limb’s position is dictated. Continuous perturbations and closed-loop system identification allow assessment of the full limb dynamics. Possibly, continuous force perturbations condition reflex adaptation by preserving natural interaction and through consistently evoking reflexive activity using continuous afferent stimulation. However, as M1 and M2 responses are blended together, neuromuscular models are necessary to extract the reflexive activity, which requires prior assumptions on linearity and what physiological components to include. The inconsistencies in the experimental findings still leave the question whether the human is able to adjust the short-latency response unanswered.
This study addressed these opposing experimental paradigms by isolating reflex contributions in the human wrist joint using the transient and the continuous approaches simultaneously in conditions where reflex modulation is expected. Subjects (n = 11) held the handle of a robotic manipulator with the forearm restrained in an arm support and performed two motor tasks in face of continuous perturbations: "maintain position, minimize position deviations" (position task, PT) and "maintain force, minimize force deviations" (force task, FT). During the continuous tasks randomly-timed transient perturbations were applied, where the continuous perturbation was briefly interrupted. Increasing damping of the environment and reduced continuous perturbation bandwidth in a PT did not affect the contraction level and M1, but did decrease the subject’s mechanical joint admittance and increase the M2. With respect to the PT, subjects increased the joint admittance and decreased M2 during the FT. Yet, counterproductive to the task and in contrast to the previous neuromuscular modelling results of decreased short-latency reflexes, the M1 increased in the FT with respect to the PT. Results from this study show adaptation of the M1 response, but indicate that the reflexive feedback from short-latency pathways obtained via the neuromuscular model does not directly map to the M1 elicited by transient perturbations.
|
11:15
15 mins
|
NEURO-MUSCULOSKELETAL SIMULATION OF INSTRUMENTED SPASTICITY ASSESSMENT IN CHILDREN WITH CEREBRAL PALSY
Marjolein van der Krogt, Lynn Bar-On, Thalia Kindt, Kaat Desloovere, Jaap Harlaar
Abstract: Increased joint resistance is an important phenomenon in patients with cerebral palsy (CP). Experimental data combined with musculoskeletal modelling may give insight in the underlying passive and neural aspects. We aimed to determine to what extent muscle response to slow and fast passive stretches can be explained by a combination of increased passive stiffness and by spasticity as defined by Lance [1], i.e. a (purely) velocity-dependent hyper-activity to passive stretch.
Experimental data was collected using an instrumented manual spasticity test [2] for 11 CP and 5 typically developing (TD) children, for slow and fast passive stretches of the hamstrings. A generic model [3] was simplified and scaled to individual subject sizes. Passive muscle properties (strain ‘s’, i.e. the stretch at an applied force equal to maximum isometric muscle force, and shape factor ‘k’) of lumped vasti and hamstrings muscles were optimized to match the slow stretch moment-angle curve. Forward dynamic (FD) simulation of slow stretches was performed to verify that the model was capable of replicating data. Next, FD simulations of fast stretches were performed, including a controller that added velocity-dependent activity to the CP muscles. This controller incorporated a velocity threshold (based on experimental data), an individually tuned gain factor and a fixed time-delay of 0.030 s. FD outcome was validated against measured knee angles and muscle activity.
The passive knee moment-angle curve was steeper in CP compared with TD and could almost be perfectly replicated by optimizing s and k (RMS error 0.10±0.06 Nm for CP and 0.22± 0.15 Nm for TD). Hamstrings strain was significantly lower in CP than in TD (0.57±0.11 vs 0.92±0.18) indicating stiffer muscles. Interestingly, all muscles except for the CP hamstrings were much more compliant than the default model’s strain of 0.60 [3]. FD simulations perfectly replicated the slow passive stretch and gave a relatively good fit for fast TD stretch. Our purely velocity-dependent spasticity model could predict muscle response to fast passive stretch in CP in terms of muscle activity, fibre length and fibre velocity. Measured knee angles were predicted well, except for the end range of motion, where sustained muscle activity (as present in most patients) was lacking in our model.
We conclude that muscle response to slow and fast passive stretches in CP and TD hamstrings could be predicted well with relatively few parameters for passive stiffness and velocity-dependent hyperactivity. In future studies, our individually tuned model can be applied to active motions such as gait.
|
11:30
15 mins
|
HUMAN FORCE REPRODUCTION ERROR DEPENDS ON FORCE DIRECTION AND ARM POSTURE
Bram Onneweer, Winfred Mugge, Alfred Schouten
Abstract: The central nervous system receives sensory information about the state of the arm and interaction forces, which allows humans to efficiently interact with objects. Proprioception provides position information through the muscle spindles and force information through the Golgi Tendon Organs and tactile sensors. While the properties of position sense have been thoroughly investigated force sense received less attention. Previous research determined the discrimination thresholds for force magnitudes and force directions. However, the influence of arm posture or length of the muscles on the force perception is yet unknown. The goal of this study is to assess the accuracy of human force reproduction in different postures (arm orientations) in the horizontal plane. Subjects (n=12, all right-handed, 8 men) were restrained in a chair and held a handle attached to a force sensor while their arm was supported to prevent shoulder ab- or adduction. They were asked to perform a series of two interchanging trials: a reference trial, where they had to match an onscreen target force (10N) in magnitude and direction with visual feedback of the magnitude and direction of the exerted force, and subsequently a reproduction trial, where they had to reproduce the same force vector without visual feedback. In each of the four postures the subjects performed the eight force directions (45 degree increments) eight times (64 trial series per posture) in random order. The first hand position was located in front of the right shoulder with the elbow in 90 degree flexion (Posture 1: shoulder flexion (SF) I, elbow flexion (EF) I). By extending the elbow to 20 degrees flexion, the second position was obtained (Posture 2: SF I, EF II). For the third hand position the shoulder was extended from the posture at the first position such that the hand was on a straight line between the centre of the body and the second hand position (Posture 3: SF II, EF I). The fourth hand position was obtained by extending the elbow of the third posture to 20 degrees flexion (Posture 4: SF II, EF II). If the force perception depends on an egocentric reference frame, the results of the second and third posture should be the same because they are in line with respect to the centre of the body. For all arm postures, ellipses were fitted through the mean force reproductions per force direction for each subject and the joint contributions were calculated using the joint to endpoint Jacobian. The least accurate direction of the force reproduction ellipse was always pointing in the direction of the shoulder for all arm postures. The joint contributions showed that force sense is most accurate in the directions where both the shoulder and elbow contributed to the generated force direction at hand level. The results of this study demonstrate that force sense depends on arm posture, which cannot be explained by an egocentric reference frame.
|
11:45
15 mins
|
NOCICEPTIVE DETECTION PROBABILITY DEPENDS ON THE TEMPORAL PROPERTIES OF ELECTROCUTANEOUS STIMULI
Robert-Jan Doll, Floor Maten, Sjoerd Spaan, Peter Veltink, Jan Buitenweg
Abstract: Electrical stimulation of the skin using a needle-electrode specifically activates nociceptive nerve fibres. The detection of such stimuli by subjects depends on the activation of subsequent nociceptive mechanisms. Activation of these mechanisms depends on the temporal stimulus properties, such as the pulse-width, number of pulses, and inter-pulse interval. This different activation of nociceptive mechanisms is reflected in nociceptive (detection) thresholds [1]. The threshold is a commonly used measure to detect nociceptive malfunctioning and is defined as the stimulus amplitude resulting in a 50% detection probability in the psychophysical curve [2]. In addition to the threshold, the psychophysical curve is also described by a slope parameter, indicating the steepness of the curve. While the effect of stimulus parameters on the threshold was studied before, the effect on the slope was not studied before. The slope could be dependent on the processing of the stimulus by the underlying mechanisms, and thus on the temporal stimulus properties. Moreover, the combination of threshold and slope estimates could possibly be used to detect malfunctioning of nociceptive mechanisms. Here, we study whether different temporal stimulus properties do not only affect the detection threshold, but also the slope of the psychophysical curve.
30 healthy human subjects were included in a 10-minute psychophysical experiment. All experimental procedures were approved by the local ethics committee. Subjects were presented with electrocutaneous stimuli with various temporal properties and were to indicate detected ones. The pulse-width (either 480 or 820 µs), number of pulses (either 1 or 2 pulses), and the inter-pulse interval (10, 50, or 100 ms) were varied in this experiment. Generalized linear mixed models with a logit link function were used to obtain estimates of the detection probability given the stimulus amplitude.
The results showed that the psychophysical function, and thus the detection probability, depends on stimulus properties. Not only the threshold, but also the slope can change when different temporal properties are chosen. Moreover, habituation of the detection probability was present for all stimulus parameters, but the effect was strongest for 1-pulse stimuli. All this suggest that additional estimation of the psychophysical curve could be useful for improved observation of (several) nociceptive mechanisms.
REFERENCES
[1] E.M. van der Heide, J.R. Buitenweg, E. Marani, W.L.C. Rutten, “Single pulse and train modulation of cutaneous electrical stimulation: a comparison of methods,” J Clin Neuro-physiol, vol. 26, pp. 54-60, 2009
[2] B. Treutwein, “Adaptive psychophysical procedures,” Vision Research, 35(17), 2503-2522
|
|
