Showing papers by "Christian Cipriani published in 2020"

Neural feedback strategies to improve grasping coordination in neuromusculoskeletal prostheses

Abstract: Conventional prosthetic arms suffer from poor controllability and lack of sensory feedback. Owing to the absence of tactile sensory information, prosthetic users must rely on incidental visual and auditory cues. In this study, we investigated the effect of providing tactile perception on motor coordination during routine grasping and grasping under uncertainty. Three transhumeral amputees were implanted with an osseointegrated percutaneous implant system for direct skeletal attachment and bidirectional communication with implanted neuromuscular electrodes. This neuromusculoskeletal prosthesis is a novel concept of artificial limb replacement that allows to extract control signals from electrodes implanted on viable muscle tissue, and to stimulate severed afferent nerve fibers to provide somatosensory feedback. Subjects received tactile feedback using three biologically inspired stimulation paradigms while performing a pick and lift test. The grasped object was instrumented to record grasping and lifting forces and its weight was either constant or unexpectedly changed in between trials. The results were also compared to the no-feedback control condition. Our findings confirm, in line with the neuroscientific literature, that somatosensory feedback is necessary for motor coordination during grasping. Our results also indicate that feedback is more relevant under uncertainty, and its effectiveness can be influenced by the selected neuromodulation paradigm and arguably also the prior experience of the prosthesis user.

Grasp force estimation from the transient EMG using high-density surface recordings

Abstract: Objective: Understanding the neurophysiological signals underlying voluntary motor control and decoding them for prosthesis control are among the major challenges in applied neuroscience and bioengineering. Usually, information from the electrical activity of residual forearm muscles (i.e. the electromyogram, EMG) is used to control different functions of a prosthesis. Noteworthy, forearm EMG patterns at the onset of a contraction (transient phase) have shown to contain predictive information about upcoming grasps. However, decoding this information for the estimation of grasp force was so far overlooked. Approach: High Density-EMG signals (192 channels) were recorded from twelve participants performing a pick-and-lift task. The final grasp force was estimated offline using linear regressors, with four subsets of channels and ten features obtained using three channels-features selection methods. Two different evaluation metrics (absolute error and R2), complemented with statistical analysis, were used to select the optimal configuration of the parameters. Different windows of data starting at the grasp force (GF) onset were compared to determine the time at which the grasp force can be ascertained from the EMG signals. Main results: The prediction accuracy improved by increasing the window length from the moment of the onset and kept improving until the steady state at which a plateau of performances was reached. With our methodology, estimations of the grasp force through 16 EMG channels reached an absolute error of 2.52% the maximum voluntary force using only transient information and 1.99% with the first 500ms of data following the onset. Significance: The final GF estimation from transient EMG was comparable to the one obtained using steady state data, confirming our hypothesis that the transient phase contains information about the final grasp force. This result paves the way to fast online myoelectric controllers capable of decoding grasp strength from the very early portion of the EMG signal.

Online Grasp Force Estimation From the Transient EMG

Abstract: Myoelectric upper limb prostheses are controlled using information from the electrical activity of residual muscles (i.e. the electromyogram, EMG). EMG patterns at the onset of a contraction (transient phase) have shown predictive information about upcoming grasps. However, decoding this information for the estimation of the grasp force was so far overlooked. In a previous offline study, we proved that the transient phase of the EMG indeed contains information about the grasp force and determined the best algorithm to extract this information. Here we translated those findings into an online platform to be tested with both non-amputees and amputees. The platform was tested during a pick and lift task (tri-digital grasp) with light objects (200 g – 1 kg), for which fine control of the grasp force is more important. Results show that, during this task, it is possible to estimate the target grasp force with an absolute error of 2.06 (1.32) % and 2.04 (0.49) % the maximum voluntary force for non-amputee and amputees, respectively, using information from the transient phase of the EMG. This approach would allow for a biomimetic regulation of the grasp force of a prosthetic hand. Indeed, the users could contract their muscles only once before the grasp begins with no need to modulate the grasp force for the whole duration of the grasp, as required with continuous classifiers. These results pave the way to fast, intuitive and robust myoelectric controllers of limb prostheses.

Feasibility of Tracking Multiple Implanted Magnets With a Myokinetic Control Interface: Simulation and Experimental Evidence Based on the Point Dipole Model

Abstract: Objective: The quest for an intuitive and physiologically appropriate human-machine interface for the control of dexterous prostheses is far from being completed. To control a hand prosthesis, a possible approach could consist in using information related to the displacement of forearm muscles of an amputee during contraction. We recently proposed that muscle displacement could be monitored by implanting passive magnetic markers (MMs– i.e., permanent magnets) in them. We dubbed this the myokinetic interface . However, besides the system feasibility, how much its accuracy, precision and computation time are affected by the number and distribution of both the MMs and the sensors used to record the MF was not quantified. Methods: Here we investigated, through simulations validated with a physical system, the performance of a system capable to track position and orientation of up to 9 MMs using information from up to 112 sensors in a volume resembling the dimensions of the human forearm. Results: The system was able to track up to 7 MMs in 450 ms, demonstrating position/orientation accuracies in the range of 1 mm/5°. The comparison with the experimental recordings demonstrated a median difference with the simulations in the order of 0.45 mm. Conclusion: We were able to formulate general guidelines for the implementation of magnetic tracking systems. Significance: Our results pave the way towards the development of new human-machine interfaces for the control of artificial limbs, but they are also interesting for the whole range of biomedical engineering applications exploiting magnetic tracking.

A database of multi-channel intramuscular electromyogram signals during isometric hand muscles contractions

Abstract: Hand movement is controlled by a large number of muscles acting on multiple joints in the hand and forearm. In a forearm amputee the control of a hand prosthesis is traditionally depending on electromyography from the remaining forearm muscles. Technical improvements have made it possible to safely and routinely implant electrodes inside the muscles and record high-quality signals from individual muscles. In this study, we present a database of intramuscular EMG signals recorded with fine-wire electrodes alongside recordings of hand forces in an isometric setup and with the addition of spike-sorted metadata. Six forearm muscles were recorded from twelve able-bodied subjects and nine forearm muscles from two subjects. The fully automated recording protocol, based on command cues, comprised a variety of hand movements, including some requiring slowly increasing/decreasing force. The recorded data can be used to develop and test algorithms for control of a prosthetic hand. Assessment of the signals was done in both quantitative and qualitative manners.

‘Doublecheck: a sensory confirmation is required to own a robotic hand, sending a command to feel in charge of it’

Abstract: Over a lifetime of experience, the representation of the body is built upon congruent integration of multiple elements constituting the sensorimotor loop. To investigate its robustness against the rupture of congruency between senses and with motor command, we selectively manipulated in healthy subjects the binds between sight, proprioception, and efferent motor command. Two experiments based on the Moving Hand Illusion were designed employing Tendon Vibration Illusion to modulate proprioception and generate illusory altered feedback of movement. In Experiment A, visuomotor congruency was modulated by introducing adelay between complex multifingered movements performed by arobotic hand and real movement of each participant's hand. In the presence of the motor command, visuomotor congruency enhanced ownership, agency, and skin conductance, while proprioceptive-motor congruency was not effective, confirming the prevalence of vision upon proprioception. In Experiment B, the impact of visuo-proprioceptive congruency was tested in the absence of motor command because the robotic hand moved autonomously. Intersensory congruency compensated for the absence of motor command only for ownership. Skin conductance in Exp Band Proprioceptive Drift in both experiments did not change. Results suggest that ownership and agency are independently processed, and presence of the efferent component modulates sensory feedbacks salience. The brain seems to require the integration of at least two streams of congruent information. Bodily awareness can be generated from sensory information alone, but to feel in charge of the body, senses must be double-checked with the prediction generated from efference copy, which is treated as an additional sensory modality.

Comparison of online algorithms for the tracking of multiple magnetic targets in a myokinetic control interface

Abstract: Magnetic tracking algorithms can be used to determine the position and orientation of magnetic markers or devices. These techniques are particularly interesting for biomedical applications such as teleoperated surgical robots or the control of upper limb prostheses. The performance of different algorithms used for magnetic tracking was compared in the past. However, in most cases, those algorithms were required to track a single magnet.Here we investigated the performance of three localization algorithms in tracking up to 9 magnets: two optimization-based (Levenberg-Marquardt algorithm, LMA, and Trust Region Reflective algorithm, TRRA) and one recursion-based (Unscented Kalman Filter, UKF). The tracking accuracy of the algorithms and their computation time were investigated through simulations.The accuracy of the three algorithms, when tracking up to six magnets, was similar, leading to estimation errors varying from 0.06 ± 0.02 mm to 2.26 ± 0.07 mm within a 100 mm × 54 mm × 100 mm workspace, at the highest sampling frequency. In all cases, computation times under 300 ms for the UKF and 45 ms for the LMA/TRRA were obtained. The TRRA showed the best tracking performance overall.These outcomes are of interest for a wide range of robotics applications that require remote tracking.

The Myokinetic Control Interface: How Many Magnets Can be Implanted in an Amputated Forearm? Evidence From a Simulated Environment

Abstract: We recently introduced the concept of a new human-machine interface (the myokinetic control interface ) to control hand prostheses. The interface tracks muscle contractions via permanent magnets implanted in the muscles and magnetic field sensors hosted in the prosthetic socket. Previously we showed the feasibility of localizing several magnets in non-realistic workspaces. Here, aided by a 3D CAD model of the forearm, we computed the localization accuracy simulated for three different below-elbow amputation levels, following general guidelines identified in early work. To this aim we first identified the number of magnets that could fit and be tracked in a proximal (T1), middle (T2) and distal (T3) representative amputation, starting from 18, 20 and 23 eligible muscles, respectively. Then we ran a localization algorithm to estimate the poses of the magnets based on the sensor readings. A sensor selection strategy (from an initial grid of 840 sensors) was also implemented to optimize the computational cost of the localization process. Results showed that the localizer was able to accurately track up to 11 (T1), 13 (T2) and 19 (T3) magnetic markers (MMs) with an array of 154, 205 and 260 sensors, respectively. Localization errors lower than 7% the trajectory travelled by the magnets during muscle contraction were always achieved. This work not only answers the question: “how many magnets could be implanted in a forearm and successfully tracked with a the myokinetic control approach?”, but also provides interesting insights for a wide range of bioengineering applications exploiting magnetic tracking.

Continuous supplementary tactile feedback can be applied (and then removed) to enhance precision manipulation

Abstract: Human sensorimotor control of dexterous manipulation relies on afferent sensory signals. Explicit tactile feedback is generally not available to prosthetic hand users, who have to rely on incidental information sources to partly close the control loop, resulting in suboptimal performance and manipulation difficulty. Recent studies on non-invasive supplementary sensory feedback indicated that time-discrete vibrational feedback delivered upon relevant mechanical events outperforms continuous tactile feedback. However, we hypothesize that continuous tactile feedback can be more effective in non-routine manipulation tasks (i.e., tasks where the grip force is modified reactively in response to the sensory feedback due to the unpredictable behavior of the manipulated object, such as picking and holding a virtual fragile object) if delivered to highly sensitive areas. We further hypothesize that this continuous tactile feedback is not necessary during all the duration of the manipulation task, since adaptation occurs. We investigated the effectiveness of continuous tactile feedback in precision manipulation, together with a new sensory feedback policy, where the continuous tactile feedback is gradually removed when the grasp reaches a steady state (namely, transient tactile feedback). We carried out an experiment in a virtual-reality setting with custom tactile feedback devices, which can apply continuous pressure and vibrations, attached to the thumb and index finger. We enrolled 24 healthy participants and instructed them to pick and hold a fragile virtual cube without breaking it. We compared their manipulation performance when using four different sensory feedback methods, i.e., no tactile feedback, discrete vibrations, continuous tactile feedback, and transient tactile feedback. The latter consisted of gradually removing the continuous feedback in the static phase of the grasp. Continuous tactile feedback leads to a significantly larger number of successful trials than discrete vibrational cues and no feedback conditions, yet the gradual removal of the continuous feedback yields to comparable outcomes. Moreover, the participants preferred the continuous stimuli over the vibrational cues and the removal in the static phase did not significantly impact their appreciation of the continuous tactile feedback. These results advocate for the use of continuous supplementary tactile feedback for fine manipulation control and indicate that it can seamlessly be removed in the static phase of the grasp, possibly due to the mechanism of sensory adaptation. This encourages the development of energy-efficient supplementary feedback devices for prosthetic and telemanipulation applications, where encumbrance and power consumption are burdensome constraints.

Discrimination of Object Curvature Based on a Sparse Tactile Sensor Array

Abstract: Object curvature plays an important role in grasping and manipulation. To be more exact, local curvature is a more useful information for grasping practically. Vision and touch are the two main methods to extract surface curvature of an object, but vision is often limited since the complete contact area is invisible during manipulation. In this paper, the authors propose an object curvature estimation method based on an artificial neural network algorithm through a lab-developed sparse tactile sensor array. The compliant layer covering on the sensor is indispensable for fitting the curved surface. Three types (plane, convex sphere, and convex cylinder) of sample and each type of sample including 30 different radiuses (1 mm to 30 mm) were used in the experiment. The overall classification accuracy was 93.1%. The average curvature radius estimating error based on an artificial neural network (ANN) algorithm was 1.87 mm. When the radius of curvature was bigger than 5 mm, the average relative error was smaller than 20%. As a comparison, the sensor array density we used in this paper was less than 9/cm2, which was smaller than the density of human SAII receptors, but the discrimination result was close to the SAII receptors. Comparison with the curvature discrimination ability of the human body showed that this method has a promising application prospect.

A Parallel Actuated Haptic Device for De-localized Tactile Feedback in Prosthetics

Abstract: This work presents a parallel actuated haptic device designed for delivering de-localized tactile feedback in prosthetics. Its peculiar design seeks to improve balance between static and slow output forces, such as grip force, and wide bandwidth tactile signals, such as vibrations, textures and fast contact transients. To this aim the device implements two actuators coupled in parallel with different mechanical transmissions. A cantilever mechanism obtains a compact arrangement of the moving parts and improves wearability of the device. We present results of the preliminary tests of the device using a force sensor: they show the enhanced capabilities of the device in modulation of the output force from static to high frequency components.

Myokinetic prosthesis control oriented environmental magnetic disturb analysis

Abstract: Several medical applications involve the use of remote magnet tracking for retrieving the position of tools instrumented with one or more magnets, when a free line-of-sight between the magnets and the tracker is not available. Our group recently proposed to implant passive magnetic markers (i.e. permanent magnets) in the forearm muscles of an amputee in order to track the displacements of those muscles during contraction. The idea is to use the retrieved information to control a hand prosthesis. We called this the myokinetic control interface. However, besides the system feasibility, how much its accuracy and precision are affected by external noise sources has not been quantified yet.Here, through an experimental setup, we investigated the influence of different magnetic/electromagnetic interferences on the localization accuracy of three permanent magnets. The magnetic field generated by the magnets was collected both in interference-free conditions and in presence of disturbances. Localization errors achieved under different conditions, and for both raw and low-pass filtered signals, were derived. Results showed that the steel bar caused the maximum average localization error, equal to 9.8 mm and 74° in terms of position and orientation, respectively. The microwave oven caused instead the maximum localization variability, with a standard deviation of 0.21 mm and 2.2°. The low-pass filtering operation (5 Hz cut-off frequency) did not lead to significant improvement in the accuracy, resulting in an error decrease always below 7% compared to the unfiltered signals.This work is important because it gives a quantitative measure of the disturbances encountered in everyday life which could cause the failure of those systems exploiting remote tracking.

The myokinetic control interface: how many magnets can be implanted in an amputated forearm? evidences from a simulated environment

Abstract: The displacement of residual muscles during voluntary contraction in a transradial amputee could be effectively exploited to control multiple degrees of freedom in a hand prosthesis. We recently introduced a new human-machine interface (the myokinetic control interface) which aims at tracking muscles contraction through implanted permanent magnets and magnetic field sensors located inside the socket. Magnetic markers (MM) tracking systems have been widely investigated in the past, especially for controlling and guiding medical tools for intra-body applications. However, specific design rules for a multiarticulate robotic hand control system have not been defined yet. Here, we studied the tracking accuracy of multiple implanted magnets by simulating different levels of trans-radial amputation using a 3D CAD model of the forearm. A magnets placing procedure was developed to position the MMs in the available muscles, following general guidelines derived in our previous study. The localizer was able to accurately track up to 9, 13 and 18 MMs, in a proximal, middle and distal representative amputation, respectively. Localization errors below ~3% the length of the trajectories travelled by the MMs during muscles contraction were achieved for all amputation levels. Not only this work answers the question: “how many magnets could be implanted in a forearm and successfully tracked with a myokinetic control approach?”, but also provides interesting insights for a wide range of bioengineering applications exploiting remote tracking.

Effects of Non-in Situ Vibrations on Hand Sensations: A Pilot Study

Abstract: The use of hand prostheses implies manipulating objects without tactile information. Although prosthetic users might benefit from supplementary sensory feedback, its advantages might be hindered by unreliable feedforward control of the prosthetic devices. The ideal tool to study supplementary sensory feedback with healthy participants is the sensory deprived human hand. In this work, we propose the use of a technique to reduce tactile sensation of the fingertips without the need for local anesthesia, based on the application of a continuous vibrational stimulus to the wrist. We quantitatively investigated its effects with established clinical tests, and we obtained encouraging yet preliminary results. The extent of this sensory reduction should be explored in further experiments with larger pools of participants.

Post-stroke Voluntary Movements Improve When Combined with Vibration-Induced Illusion of Movement

Abstract: Improving upper limb motor recovery is a major goal of stroke rehabilitation. However, after rehabilitation, stroke patients often still experience difficulty completing motor tasks. Sensory feedback can be impaired by stroke damage, and inadequate feedback and integration with motor control may prevent these patients from achieving smooth motor coordination. Here, we tested the effects of augmenting natural movement proprioception with vibration-induced illusions of movement during a reaching task. When participants reached for a target with the affected arm (stroke patients) or dominant arm (able-bodied participants), 90 Hz vibration on either the biceps brachii or triceps brachii induced the illusion of movement. Participants also completed control trials where sham (25 Hz) or no vibration were applied. Fitts’ Law and kinematic analyses revealed that vibration-induced movement illusions delivered to the primary agonist muscle involved in active movement improved stroke patients’ reaching motions. Incorporating this technique in rehabilitation could promote functional motor recovery.