Prosthesis Control Research Articles

Online Classification of Transient EMG Patterns for the Control of the Wrist and Hand in a Transradial Prosthesis

Decoding human motor intentions by processing electrophysiological signals is a crucial, yet unsolved, challenge for the development of effective upper limb prostheses. Pattern recognition of continuous myoelectric (EMG) signals represents the state-of-art for multi-DoF prosthesis control. However, this approach relies on the unreliable assumption that repeatable muscular contractions produce repeatable patterns of steady-state EMGs. Here, we propose an approach for decoding wrist and hand movements by processing the signals associated with the onset of contraction (transient EMG). Specifically, we extend the concept of a transient EMG controller for the control of both wrist and hand, and tested it online. We assessed it with one transradial amputee and 15 non-amputees via the Target Achievement Control test. Non-amputees successfully completed 95% of the trials with a median completion time of 17 seconds, showing a significant learning trend ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">p</i> < 0.001). The transradial amputee completed about the 80% of the trials with a median completion time of 26 seconds. Although the performance proved comparable with earlier studies, the long completion times suggest that the current controller is not yet clinically viable. However, taken collectively, our outcomes reinforce earlier hypothesis that the transient EMG could represent a viable alternative to steady-state pattern recognition approaches.

Neural evidence for functional roles of tactile and visual feedback in the application of myoelectric prosthesis

Objective. The primary purpose of this study was to investigate the electrophysiological mechanism underlying different modalities of sensory feedback and multi-sensory integration in typical prosthesis control tasks. Approach. We recruited 15 subjects and developed a closed-loop setup for three prosthesis control tasks which covered typical activities in the practical prosthesis application, i.e. prosthesis finger position control (PFPC), equivalent grasping force control (GFC) and box and block control (BABC). All the three tasks were conducted under tactile feedback (TF), visual feedback (VF) and tactile-visual feedback (TVF), respectively, with a simultaneous electroencephalography (EEG) recording to assess the electroencephalogram (EEG) response underlying different types of feedback. Behavioral and psychophysical assessments were also administered in each feedback condition. Results. EEG results showed that VF played a predominant role in GFC and BABC tasks. It was reflected by a significantly lower somatosensory alpha event-related desynchronization (ERD) in TVF than in TF and no significant difference in visual alpha ERD between TVF and VF. In PFPC task, there was no significant difference in somatosensory alpha ERD between TF and TVF, while a significantly lower visual alpha ERD was found in TVF than in VF, indicating that TF was essential in situations related to proprioceptive position perception. Tactile-visual integration was found when TF and VF were congruently implemented, showing an obvious activation over the premotor cortex in the three tasks. Behavioral and psychophysical results were consistent with EEG evaluations. Significance. Our findings could provide neural evidence for multi-sensory integration and functional roles of tactile and VF in a practical setting of prosthesis control, shedding a multi-dimensional insight into the functional mechanisms of sensory feedback.

An LWPR-Based Method for Intelligent Lower-Limb Prosthesis Control by Learning the Dynamic Model in Real Time

A significant number of people in the world suffer from limb losses, while prosthesis is the hopeful way to help the amputees back to normal life. Recently, the most popular control method used in intelligent prosthesis is FSM-IC (finite state machine with impedance control), which requires a significant amount of manual parameter adjustments to achieve a good model compensation in a discrete way. Taking the lower-limb prosthesis as the research object, this paper applies an LWPR (locally weighted projection regression) model to learn the dynamic model of a prosthesis in real time in order to achieve a better model compensation in a continuous way and propose scientific experimental schemes to verify the control method. First, the basic control framework of lower-limb prosthesis is given. Then, the control law is derived on the basis of model building and LWPR’s addition. Finally, the proper experimental schemes are designed to carry out the control method effectively in a safe way. The experimental results show that the control law with the LWPR model can greatly improve the tracking performance during the swing phase and obtain rather good compliance during the stance phase. Moreover, the results also indicate that the LWPR model can approximate the dynamic model online. This method is hoped to be extended to more applications and fields.

Limb loading enhances skill transfer between augmented and physical reality tasks during limb loss rehabilitation

BackgroundVirtual and augmented reality (AR) have become popular modalities for training myoelectric prosthesis control with upper-limb amputees. While some systems have shown moderate success, it is unclear how well the complex motor skills learned in an AR simulation transfer to completing the same tasks in physical reality. Limb loading is a possible dimension of motor skill execution that is absent in current AR solutions that may help to increase skill transfer between the virtual and physical domains.MethodsWe implemented an immersive AR environment where individuals could operate a myoelectric virtual prosthesis to accomplish a variety of object relocation manipulations. Intact limb participants were separated into three groups, the load control (CGLD; N=4), the AR control (CGAR; N=4), and the experimental group (EG; N=4). Both the CGAR and EG completed a 5-session prosthesis training protocol in AR while the CGLD performed simple muscle training. The EG attempted manipulations in AR while undergoing limb loading. The CGAR attempted the same manipulations without loading. All participants performed the same manipulations in physical reality while operating a real prosthesis pre- and post-training. The main outcome measure was the change in the number of manipulations completed during the physical reality assessments (i.e. completion rate). Secondary outcomes included movement kinematics and visuomotor behavior.ResultsThe EG experienced a greater increase in completion rate post-training than both the CGAR and CGLD. This performance increase was accompanied by a shorter motor learning phase, the EG’s performance saturating in less sessions of AR training than the CGAR.ConclusionThe results demonstrated that limb loading plays an important role in transferring complex motor skills learned in virtual spaces to their physical reality analogs. While participants who did not receive limb loading were able to receive some functional benefit from AR training, participants who received the loading experienced a greater positive change in motor performance with their performance saturating in fewer training sessions.

Myoelectric prosthesis hand grasp control following targeted muscle reinnervation in individuals with transradial amputation.

Despite the growing availability of multifunctional prosthetic hands, users' control and overall functional abilities with these hands remain limited. The combination of pattern recognition control and targeted muscle reinnervation (TMR) surgery, an innovative technique where amputated nerves are transferred to reinnervate new muscle targets in the residual limb, has been used to improve prosthesis control of individuals with more proximal upper limb amputations (i.e., shoulder disarticulation and transhumeral amputation). The goal of this study was to determine if prosthesis hand grasp control improves following transradial TMR surgery. Eight participants were trained to use a multi-articulating hand prosthesis under myoelectric pattern recognition control. All participated in home usage trials pre- and post-TMR surgery. Upper limb outcome measures were collected following each home trial. Three outcome measures (Southampton Hand Assessment Procedure, Jebsen-Taylor Hand Function Test, and Box and Blocks Test) improved 9-12 months post-TMR surgery compared with pre-surgery measures. The Assessment of Capacity for Myoelectric Control and Activities Measure for Upper Limb Amputees outcome measures had no difference pre- and post-surgery. An offline electromyography analysis showed a decrease in grip classification error post-TMR surgery compared to pre-TMR surgery. Additionally, a majority of subjects noted qualitative improvements in their residual limb and phantom limb sensations post-TMR. The potential for TMR surgery to result in more repeatable muscle contractions, possibly due to the reduction in pain levels and/or changes to phantom limb sensations, may increase functional use of many of the clinically available dexterous prosthetic hands.

Virtual regression-based myoelectric hand-wrist prosthesis control and electrode site selection using no force feedback

Most transradial prosthesis users with conventional “Sequential” myoelectric control have two electrode sites which control one degree of freedom (DoF) at a time. Rapid EMG co-activation toggles control between DoFs (e.g., hand and wrist), providing limited function. We implemented a regression-based EMG control method which achieved simultaneous and proportional control of two DoFs in a virtual task. We automated electrode site selection using a short-duration (90 s) calibration period, without force feedback. Backward stepwise selection located the best electrodes for either six or 12 electrodes (selected from a pool of 16). We additionally studied two, 2-DoF controllers: “Intuitive” control (hand open-close and wrist pronation-supination controlled virtual target size and rotation, respectively) and “Mapping” control (wrist flexion–extension and ulnar-radial deviation controlled virtual target left–right and up-down movement, respectively). In practice, a Mapping controller would be mapped to control prosthesis hand open-close and wrist pronation-supination. Eleven able-bodied subjects and 4 limb-absent subjects completed virtual target matching tasks (fixed target moves to a new location after being “matched,” and subject immediately pursues) and fixed (static) target tasks. For all subjects, both 2-DoF controllers with 6 optimally-sited electrodes had statistically better target matching performance than Sequential control in number of matches (average of 4–7 vs 2 matches, p < 0.001) and throughput (average of 0.75–1.25 vs 0.4 bits/s, p < 0.001), but not overshoot rate and path efficiency. There were no statistical differences between 6 and 12 optimally-sited electrodes for both 2-DoF controllers. These results support the feasibility of 2-DoF simultaneous, proportional myoelectric control.

A Wearable Master–Slave Rehabilitation Robot Based on an Epidermal Array Electrode Sleeve and Multichannel Electromyography Network

Investigating electromyography (EMG) signals is vital to promote the development of both rehabilitative robots and understanding of the movement neural mechanism. Interactions between various muscle units are paramount to be measured through network analysis, aiming to reveal how information is propagated and integrated. Herein, an EMG network using an epidermal array electrode sleeve to record multichannel EMGs is constructed. Then, a master–slave rehabilitation robot by adopting the EMG network as a feature for movement intention recognition is built. The results demonstrate that the sleeve can record signals with high quality, characterized by better signal robustness and higher movement recognition performance. The different finger movements evoke the specific spatial network patterns, characterized by the dominated hub at the muscle in charge of the corresponding movement, and the proposed EMG network‐based approach consistently achieves the highest recognition accuracy. Moreover, the proposed approach also shows the relatively less influence of signal length and electrode positions on the movement recognition. Finally, the proposed robot system can achieve 98.21% ± 2.37 accuracy for online control. These results provide a novel theoretical and practical basis for neural prosthesis control and hemiplegic hand rehabilitation.

Tensor Factorization and Attention-Based CNN-LSTM Deep-Learning Architecture for Improved Classification of Missing Physiological Sensors Data

One of the essential issues for efficient control of prosthesis is the accurate classification of target movements hidden in electroencephalography (EEG) and electromyography (EMG) signals. However, in the presence of missing data in acquired signals, the classification accuracy degrades significantly as the amount of missing data increases, reducing the control performance of the prosthesis. This research proposes a framework based on tensor (multidimensional array) factorization and attention-based convolutional neural network (CNN)-long short-term memory (LSTM) deep learning (DL) for recovering missing data and performing classification of target movements, respectively. To recover missing data in tensor factorization, Canonical/Polyadic Weighted OPTimization (CP-WOPT) is employed, and its performance is compared to state-of-the-art factorization methods, whereas the performance of CNN-LSTM-attention layer (Attn) is compared to state-of-the-art machine learning and DL classifiers. Results show that CNN-LSTM-Attn obtained the mean classification accuracy of 98%, 83%, and 90% on complete (0% missing data), partially complete (10% to 50% missing data), and tensor-recovered real-world EEG and EMG data, respectively, demonstrating the applicability of the proposed framework.

Force-Moment Sensor for Prosthesis Structural Load Measurement

Measurement of prosthesis structural load, as an important way to quantify the interaction of the amputee user with the environment, may serve important purposes in the control of smart lower-limb prosthetic devices. However, the majority of existing force sensors used in protheses are developed based on strain measurement and thus may suffer from multiple issues such as weak signals and signal drifting. To address these limitations, this paper presents a novel Force-Moment Prosthesis Load Sensor (FM-PLS) to measure the axial force and bending moment in the structure of a lower-limb prosthesis. Unlike strain gauge-based force sensors, the FM-PLS is developed based on the magnetic sensing of small (millimeter-scale) deflection of an elastic element, and it may provide stronger signals that are more robust against interferences and drifting since such physical deflection is several orders of magnitude greater than the strain of a typical load-bearing structure. The design of the sensor incorporates uniquely curved supporting surfaces such that the measurement is sensitive to light load but the sensor structure is robust enough to withstand heavy load without damage. To validate the sensor performance, benchtop testing of the FM-PLS and walking experiments of a FM-PLS-embedded robotic lower-limb prosthesis were conducted. Benchtop testing results displayed good linearity and a good match to the numerical simulation results. Results from the prosthesis walking experiments showed that the sensor signals can be used to detect important gaits events such as heel strike and toe-off, facilitating the reliable motion control of lower-limb prostheses.

Hierarchical approach for fusion of electroencephalography and electromyography for predicting finger movements and kinematics using deep learning

The brain is a unique organ that performs multiple processes simultaneously, such as sensory, motor, and cognitive function. However, several neurological diseases (ataxia, dystonia, Huntington’s disease) or trauma affect the limb movement and there is no cure. Although brain-computer interfaces (BCIs) have been recently used to improve the quality of life for people with severe motor disabilities, anthropomorphic control of a prosthetic hand in upper limb rehabilitation still remains an unachieved goal. To this purpose, a hierarchical integration of neural commands to fingers was applied for execution of human hand grasping with better precision. For finger movement prediction and kinematics estimation, a neuromuscular approach was employed to establish a hierarchical synergy between electroencephalography (EEG) and electromyography (EMG). EEG, EMG and metacarpophalangeal (MCP) joint kinematics were acquired during five finger flexion movements of the human hand. EMG for five finger movements and kinematics were estimated from EEG using linear regression. A Long Short-Term Memory network (LSTM) and a random forest regressor were adjoined hierarchically for prediction of finger movements and estimation of finger kinematics from the estimated EMG. The results showed an average accuracy of 84.25 ± 0.61 % in predicting finger movements and an average minimum error of 0.318 ± 0.011 in terms of root mean squared error (RMSE) in predicting finger kinematics from EEG across six subjects and five fingers. These findings suggest the implementation of a hierarchical approach to develop anthropomorphic control for upper limb prostheses.

Joint torques estimation in human gait based on Gaussian process.

Human gait involves activities in nervous and musculoskeletal dynamics to modulate joint torques with time continuously for adapting to varieties of walking conditions. The goal of this paper is to estimate the joint torques of lower limbs in human gait based on Gaussian process. The potential uses of this study include optimization of exoskeleton assistance, control of the active prostheses, and modulating the joint torque for human-like robots. To achieve this, Gaussian process (GP) based data fusion algorithm is established with joint angles as the inputs. The statistic nature of the proposed model can explore the correlations between joint angles and joint torques, and enable accurate joint-torque estimations. Experiments were conducted for 5 subjects at three walking speed (0.8 m/s, 1.2 m/s, 1.6m/s). The results show that it is possible to estimate the joint torques at different scenarios.

A Novel Framework to Facilitate User Preferred Tuning for a Robotic Knee Prosthesis.

The tuning of robotic prosthesis control is essential to provide personalized assistance to individual prosthesis users. Emerging automatic tuning algorithms have shown promise to ease the device personalization procedure. However, very few automatic tuning algorithms consider the user preference as the tuning goal, which may limit the adoptability of the robotic prosthesis. In this study, we propose and evaluate a novel prosthesis control tuning framework for a robotic knee prosthesis, which could enable user preferred robot behavior in the device tuning process. The framework consists of 1) a User-Controlled Interface that allows the user to select their preferred knee kinematics in gait and 2) a reinforcement learning-based algorithm for tuning high-dimension prosthesis control parameters to meet the desired knee kinematics. We evaluated the performance of the framework along with usability of the developed user interface. In addition, we used the developed framework to investigate whether amputee users can exhibit a preference between different profiles during walking and whether they can differentiate between their preferred profile and other profiles when blinded. The results showed effectiveness of our developed framework in tuning 12 robotic knee prosthesis control parameters while meeting the user-selected knee kinematics. A blinded comparative study showed that users can accurately and consistently identify their preferred prosthetic control knee profile. Further, we preliminarily examined gait biomechanics of the prosthesis users when walking with different prosthesis control and did not find clear difference between walking with preferred prosthesis control and when walking with normative gait control parameters. This study may inform future translation of this novel prosthesis tuning framework for home or clinical use.

Closed-Loop Control of a Multifunctional Myoelectric Prosthesis With Full-State Anatomically Congruent Electrotactile Feedback.

State-of-the-art myoelectric hand prostheses provide multi-functional control but lack somatosensory feedback. To accommodate the full functionality of a dexterous prosthesis, the artificial sensory feedback needs to convey several degrees of freedom (DoF) simultaneously. However, this is a challenge with current methods as they are characterized by a low information bandwidth. In this study, we leverage the flexibility of a recently developed system for simultaneous electrotactile stimulation and electromyography (EMG) recording to present the first solution for closed-loop myoelectric control of a multifunctional prosthesis with full-state anatomically congruent electrotactile feedback. The novel feedback scheme (coupled encoding) conveyed proprioceptive (hand aperture, wrist rotation) and exteroceptive information (grasping force). The coupled encoding was compared to the conventional approach (sectorized encoding) and incidental feedback in 10 non-disabled and one amputee participant who used the system to perform a functional task. The results showed that both feedback approaches increased the accuracy of position control compared to incidental feedback. However, the feedback increased completion time, and it did not significantly improve grasping force control. Importantly, the performance of the coupled feedback was not significantly different compared to the conventional scheme, despite the latter being easier to learn during training. Overall, the results indicate that the developed feedback can improve prosthesis control across multiple DoFs but they also highlight the subjects' ability to exploit minimal incidental information. Importantly, the current setup is the first to convey three feedback variables simultaneously using electrotactile stimulation while providing multi-DoF myoelectric control with all hardware components mounted on the same forearm.

Real-Time Wrist Motion Decoding with High Framerate Electrical Impedance Tomography (EIT).

Human wrist motion decoding with a biological-signal-based interface is a key technique in the upper-limb exoskeleton and prosthesis control. One critical issue in this field is achieving high recognition precision and fast time response while against external disturbances of sensor re-wearing. In this study, we proposed a high-framerate Electrical Impedance Tomography (EIT) system combined with an adaptive recognition algorithm for real-time wrist kinematics decoding. The high-framerate EIT system was developed by a parallel stimulation-measurement sequence, and the sampling rate was as high as 104 Hz. Compared to the most widely used myoelectric techniques, the EIT-based interface can provide extra deep muscular spatial information with similar surface electrodes. It greatly benefited the subsequent recognition algorithms, in which the key EIT regions indicating muscle morphology kept consistent after an arbitrary sensor re-donning. The designed adaptive algorithm achieved equally high performance with an automatic update of the classifier mean values with a fast self-operated calibration process. We validated the approach on 12 subjects with a 2-dimensional Fitts' law test. The wrist gestures and joint angles were mapped to the direction and speed of the cursor movement, respectively. The average throughputs (TPs) of Fitts' law tests were 1.0269 ± 0.0971 bits/s and 1.0095 ± 0.0931 bits/s without and with sensor re-donning, respectively, which were comparable to the TPs of sEMG-based studies. The results showed the promise of the EIT-based interface on real-time human motion intent recognition. Future endeavors are worth being paid in this direction for more complicated robotic tasks.

Finite-State Impedance and Direct Myoelectric Control for Robotic Ankle Prostheses: Comparing Their Performance and Exploring Their Combination.

Non-volitional control, such as finite-state machine (FSM) impedance control, does not directly incorporate user intent signals, while volitional control, like direct myoelectric control (DMC), relies on these signals. This paper compares the performance, capabilities, and perception of FSM impedance control to DMC on a robotic prosthesis for subjects with and without transtibial amputation. It then explores, using the same metrics, the feasibility and performance of the combination of FSM impedance control and DMC across the full gait cycle, termed Hybrid Volitional Control (HVC). After calibration and acclimation with each controller, subjects walked for two minutes, explored the control capabilities, and completed a questionnaire. FSM impedance control produced larger average peak torque (1.15 Nm/kg) and power (2.05 W/kg) than DMC (0.88 Nm/kg and 0.94 W/kg). The discrete FSM, however, caused non-standard kinetic and kinematic trajectories, while DMC yielded trajectories qualitatively more similar to able-bodied biomechanics. While walking with HVC, all subjects successfully achieved ankle push-off and were able to modulate the magnitude of push-off via the volitional input. Unexpectedly, however, HVC behaved either more similarly to FSM impedance control or to DMC alone, rather than in combination. Both DMC and HVC, but not FSM impedance control, allowed subjects to achieve unique activities such as tip-toe standing, foot tapping, side-stepping, and backward walking. Able-bodied subject (N=6) preferences were split amongst the controllers, while all transtibial subjects (N=3) preferred DMC. Desired performance and ease of use showed the highest correlations with overall satisfaction (0.81 and 0.82, respectively).

User Performance With a Transradial Multi-Articulating Hand Prosthesis During Pattern Recognition and Direct Control Home Use.

With the increasing availability of more advanced prostheses individuals with a transradial amputation can now be fit with single to multi-degree of freedom hands. Reliable and accurate control of these multi-grip hands still remains challenging. This is the first multi-user study to investigate at-home control and use of a multi-grip hand prosthesis under pattern recognition and direct control. Individuals with a transradial amputation were fitted with and trained to use an OSSUR i-Limb Ultra Revolution with Coapt COMPLETE CONTROL system. They participated in two 8-week home trials using the hand under myoelectric direct and pattern recognition control in a randomized order. While at home, participants demonstrated broader usage of grips in pattern recognition compared to direct control. After the home trial, they showed significant improvements in the Assessment of Capacity for Myoelectric Control (ACMC) outcome measure while using pattern recognition control compared to direct control; other outcome measures showed no differences between control styles. Additionally, this study provided a unique opportunity to evaluate EMG signals during home use. Offline analysis of calibration data showed that users were 81.5% [7.1] accurate across a range of three to five grips. Although EMG signal noise was identified during some calibrations, overall EMG quality was sufficient to provide users with control performance at or better than direct control.

A reduced-order closed-loop hybrid dynamic model for design and development of lower limb prostheses.

This manuscript presents a simplified dynamic human-prosthesis model and simulation framework for the purpose of designing and developing lower limb prosthesis hardware and controllers. The objective was to provide an offline design tool to verify the closed-loop behavior of the prosthesis with the human, in order to avoid relying solely on limiting kinematic and kinetic reference trajectories of (able-bodied) subjects and associated static or inverse dynamic analyses, while not having to resort to complete neuromusculoskeletal models of the human that require extensive optimizations to run. The presented approach employs a reduced-order model that includes only the prosthetic limb and trunk in a multi-body dynamic model. External forces are applied to the trunk during stance phase of the intact leg to represent its presence. Walking is realized by employing the well-known spring-loaded inverted pendulum model, which is shown to generate realistic dynamics on the prosthesis while maintaining a stable and modifiable gait. This simple approach is inspired from the rationale that the human is adaptive, and from the desire to facilitate modifications or inclusions of additional user actions. The presented framework is validated with two use cases, featuring a commercial and research knee prosthesis in combination with a passive ankle prosthesis, performing a continuous sequence of standing still, walking at different velocities and stopping.

Shared Autonomy Locomotion Synthesis With a Virtual Powered Prosthetic Ankle.

Virtual environments provide a safe and accessible way to test innovative technologies for controlling wearable robotic devices. However, to simulate devices that support walking, such as powered prosthetic legs, it is not enough to model the hardware without its user. Predictive locomotion synthesizers can generate the movements of a virtual user, with whom the simulated device can be trained or evaluated. We implemented a Deep Reinforcement Learning based motion controller in the MuJoCo physics engine, where autonomy over the humanoid model was shared between the simulated user and the control policy of an active prosthesis. Despite not optimising the controller to match experimental dynamics, realistic torque profiles and ground reaction force curves were produced by the agent. A data-driven and continuous representation of user intent was used to simulate a Human Machine Interface that controlled a transtibial prosthesis in a non-steady state walking setting. The continuous intent representation was shown to mitigate the need for compensatory gait patterns from their virtual users and halve the rate of tripping. Co-adaptation was identified as a potential challenge for training human-in-the-loop prosthesis control policies. The proposed framework outlines a way to explore the complex design space of robot-assisted gait, promoting the transfer of the next generation of intent driven controllers from the lab to real-life scenarios.

A Locomotion Mode Recognition Algorithm using adaptive Dynamic Movement Primitives.

Control systems of robotic prostheses should be designed to decode the users' intent to start, stop, or change locomotion; and to select the suitable control strategy, accordingly. This paper describes a locomotion mode recognition algorithm based on adaptive Dynamic Movement Primitive models used as locomotion templates. The models take foot-ground contact information and thigh roll angle, measured by an inertial measurement unit, for generating continuous model variables to extract features for a set of Support Vector Machines. The proposed algorithm was tested offline on data acquired from 10 intact subjects and 1 subject with transtibial amputation, in ground-level walking and stair ascending/descending activities. Following subject-specific training, results on intact subjects showed that the algorithm can classify initiatory and steady-state steps with up to 100.00% median accuracy medially at 28.45% and 27.40% of the swing phase, respectively. While the transitory steps were classified with up to 87.30% median accuracy medially at 90.54% of the swing phase. Results with data of the transtibial amputee showed that the algorithm classified initiatory, steady-state, and transitory steps with up to 92.59%, 100%, and 93.10% median accuracies medially at 19.48%, 51.47%, and 93.33% of the swing phase, respectively. The results support the feasibility of this approach in robotic prosthesis control.

A-Mode Ultrasound-Based Prediction of Transfemoral Amputee Prosthesis Walking Kinematics Via an Artificial Neural Network.

Lower-limb powered prostheses can provide users with volitional control of ambulation. To accomplish this goal, they require a sensing modality that reliably interprets user intention to move. Surface electromyography (EMG) has been previously proposed to measure muscle excitation and provide volitional control to upper- and lower-limb powered prosthesis users. Unfortunately, EMG suffers from a low signal to noise ratio and crosstalk between neighboring muscles, often limiting the performance of EMG-based controllers. Ultrasound has been shown to have better resolution and specificity than surface EMG. However, this technology has yet to be integrated into lower-limb prostheses. Here we show that A-mode ultrasound sensing can reliably predict the prosthesis walking kinematics of individuals with a transfemoral amputation. Ultrasound features from the residual limb of 9 transfemoral amputee subjects were recorded with A-mode ultrasound during walking with their passive prosthesis. The ultrasound features were mapped to joint kinematics through a regression neural network. Testing of the trained model against untrained kinematics from an altered walking speed show accurate predictions of knee position, knee velocity, ankle position, and ankle velocity, with a normalized RMSE of 9.0 ± 3.1%, 7.3 ± 1.6%, 8.3 ± 2.3%, and 10.0 ± 2.5% respectively. This ultrasound-based prediction suggests that A-mode ultrasound is a viable sensing technology for recognizing user intent. This study is the first necessary step towards implementation of volitional prosthesis controller based on A-mode ultrasound for individuals with transfemoral amputation.