Offline Evaluation Matters: Investigation of the Influence of Offline Performance on Real-Time Operation of Electromyography-Based Neural-Machine Interfaces

Screen Guided Training Does Not Capture Goal-Oriented Behaviours: Learning Myoelectric Control Mappings From Scratch Using Context Informed Incremental Learning

Human-machine interfaces based on myoelectric signals typically use screen-guided training (SGT) for model calibration, but this approach fails to capture realistic user behaviors. This study evaluates a user-in-the-loop context-informed incremental learning (CIIL) framework, comparing SGT, SGT followed by CIIL adaptation (SGT-A), and a novel zero-shot adaptation (ZS-A) CIIL approach that begins adapting with no prior training. Sixteen participants completed a Fitts' Law targeting task using these control schemes, with performance measured via online throughput and offline classification accuracy. Despite lower offline accuracy, the ZS-A model achieved the highest online throughput (1.47 ± 0.46 bits/s), significantly outperforming the SGT baseline (1.15 ± 0.37 bits/s) and reached competitive performance within 200 seconds. To further enhance control performance, a novel adaptive sigmoid-based proportional control mapping was introduced, dynamically adjusting control signals to allow precise control near neutral positions and rapid movements at higher activation levels, better aligning with natural user behaviors. These findings demonstrate that CIIL can surpass traditional SGT methods in online performance and emphasize the value of real-time user-in-the-loop data for developing adaptable and intuitive myoelectric interfaces, with implications for prosthetics, rehabilitation, and telerobotics.

An Efficient Framework for Personalizing EMG-Driven Musculoskeletal Models Based on Reinforcement Learning

This study aimed to develop a novel framework to quickly personalize electromyography (EMG)-driven musculoskeletal models (MMs) as efferent neural interfaces for upper limb prostheses. Our framework adopts a generic upper-limb MM as a baseline and uses an artificial neural network-based policy to fine-tune the model parameters for MM personalization. The policy was trained by reinforcement learning (RL) to heuristically adjust the MM parameters to maximize the accuracy of estimated hand and wrist motions from EMG inputs. Our present framework was compared to the baseline MM and a widely used MM parameter optimization method: simulated annealing (SA). An offline evaluation was performed to first quantify the time required for MM personalization and the kinematics estimation accuracy of personalized MMs based on data collected from non-disabled subjects. Then, in an online evaluation, additional human subjects, including an individual with a transradial amputation, performed a virtual hand posture matching task using generic and personalized MMs. Results showed that compared to the baseline generic MM, personalized MMs estimated joint motion with lower error in both offline ( ) and online tests ( ), demonstrating the benefit of MM personalization. The RL-based framework performed model optimization in under one second on average in cases that took SA over 13 minutes and yielded comparable kinematics estimations both offline and online. Hence, our present personalization framework can be a practical solution for the daily use of EMG-driven MMs in prostheses or other assistive devices.

Integrating Upper-Limb Prostheses with the Human Body: Technology Advances, Readiness, and Roles in Human-Prosthesis Interaction

Significant advances in bionic prosthetics have occurred in the past two decades. The field's rapid expansion has yielded many exciting technologies that can enhance the physical, functional, and cognitive integration of a prosthetic limb with a human. We review advances in the engineering of prosthetic devices and their interfaces with the human nervous system, as well as various surgical techniques for altering human neuromusculoskeletal systems for seamless human-prosthesis integration. We discuss significant advancements in research and clinical translation, focusing on upper limbprosthetics since they heavily rely on user intent for daily operation, although many discussed technologies have been extended to lower limb prostheses as well. In addition, our review emphasizes the roles of advanced prosthetics technologies in complex interactions with humans and the technology readiness levels (TRLs) of individual research advances. Finally, we discuss current gaps and controversies in the field and point out future research directions, guided by TRLs.

Understanding the influence of confounding factors in myoelectric control for discrete gesture recognition

Discrete myoelectric control-based gesture recognition has recently gained interest as a possible input modality for many emerging ubiquitous computing applications. Unlike the continuous control commonly employed in powered prostheses, discrete systems seek to recognize the dynamic sequences associated with gestures to generate event-based inputs. More akin to those used in general-purpose human-computer interaction, these could include, for example, a flick of the wrist to dismiss a phone call or a double tap of the index finger and thumb to silence an alarm. Moelectric control systems have been shown to achieve near-perfect classification accuracy, but in highly constrained offline settings. Real-world, online systems are subject to 'confounding factors' (i.e. factors that hinder the real-world robustness of myoelectric control that are not accounted for during typical offline analyses), which inevitably degrade system performance, limiting their practical use. Although these factors have been widely studied in continuous prosthesis control, there has been little exploration of their impacts on discrete myoelectric control systems for emerging applications and use cases. Correspondingly, this work examines, for the first time, three confounding factors and their effect on the robustness of discrete myoelectric control: (1)limb position variability, (2)cross-day use, and a newly identified confound faced by discrete systems (3)gesture elicitation speed. Results from four different discrete myoelectric control architectures: (1) Majority Vote LDA, (2) Dynamic Time Warping, (3) an LSTM network trained with Cross Entropy, and (4) an LSTM network trained with Contrastive Learning, show that classification accuracy is significantly degraded (p<0.05) as a result of each of these confounds. This work establishes that confounding factors are a critical barrier that must be addressed to enable the real-world adoption of discrete myoelectric control for robust and reliable gesture recognition.

Context-informed incremental learning improves both the performance and resilience of myoelectric control

Despite its rich history of success in controlling powered prostheses and emerging commercial interests in ubiquitous computing, myoelectric control continues to suffer from a lack of robustness. In particular, EMG-based systems often degrade over prolonged use resulting in tedious recalibration sessions, user frustration, and device abandonment. Unsupervised adaptation is one proposed solution that updates a model's parameters over time based on its own predictions during real-time use to maintain robustness without requiring additional user input or dedicated recalibration. However, these strategies can actually accelerate performance deterioration when they begin to classify (and thus adapt) incorrectly, defeating their own purpose. To overcome these limitations, we propose a novel adaptive learning strategy, Context-Informed Incremental Learning (CIIL), that leverages in situ context to better inform the prediction of pseudo-labels. In this work, we evaluate these CIIL strategies in an online target acquisition task for two use cases: (1) when there is a lack of training data and (2) when a drastic and enduring alteration in the input space has occurred. A total of 32 participants were evaluated across the two experiments. The results show that the CIIL strategies significantly outperform the current state-of-the-art unsupervised high-confidence adaptation and outperform models trained with the conventional screen-guided training approach, even after a 45-degree electrode shift (p < 0.05). Consequently, CIIL has substantial implications for the future of myoelectric control, potentially reducing the training burden while bolstering model robustness, and leading to improved real-time control.

Robust myoelectric pattern recognition methods for reducing users' calibration burden: challenges and future

Myoelectric pattern recognition (MPR) has evolved into a sophisticated technology widely employed in controlling myoelectric interface (MI) devices like prosthetic and orthotic robots. Current MIs not only enable multi-degree-of-freedom control of prosthetic limbs but also demonstrate substantial potential in consumer electronics. However, the non-stationary random characteristics of myoelectric signals poses challenges, leading to performance degradation in practical scenarios such as electrode shifting and switching new users. Conventional MIs often necessitate meticulous calibration, imposing a significant burden on users. To address user frustration during the calibration process, researchers have focused on identifying MPR methods that alleviate this burden. This article categorizes common scenarios that incur calibration burdens as based on data distribution shift and based on dynamic data categories. Then further investigated and summarized the popular robust MPR algorithms used to reduce the user's calibration burden. We categorize these algorithms as based on data manipulate, feature manipulation and, model structure. And describes the scenarios to which each method is applicable and the conditions required for calibration. Finally, this review is concluded with the advantages of robust MPR and the remaining challenges and future opportunities.

Ankle Torque Estimation With Motor Unit Discharges in Residual Muscles Following Lower-Limb Amputation

There has been increased interest in using residual muscle activity for neural control of powered lower-limb prostheses. However, only surface electromyography (EMG)-based decoders have been investigated. This study aims to investigate the potential of using motor unit (MU)-based decoding methods as an alternative to EMG-based intent recognition for ankle torque estimation. Eight people without amputation (NON) and seven people with amputation (AMP) participated in the experiments. Subjects conducted isometric dorsi- and plantarflexion with their intact limb by tracing desired muscle activity of the tibialis anterior (TA) and gastrocnemius (GA) while ankle torque was recorded. To match phantom limb and intact limb activity, AMP mirrored muscle activation with their residual TA and GA. We compared neuromuscular decoders (linear regression) for ankle joint torque estimation based on 1) EMG amplitude (aEMG), 2) MU firing frequencies representing neural drive (ND), and 3) MU firings convolved with modeled twitch forces (MUDrive). In addition, sensitivity analysis and dimensionality reduction of optimization were performed on the MUDrive method to further improve its practical value. Our results suggest MUDrive significantly outperforms (lower root-mean-square error) EMG and ND methods in muscles of NON, as well as both intact and residual muscles of AMP. Reducing the number of optimized MUDrive parameters degraded performance. Even so, optimization computational time was reduced and MUDrive still outperformed aEMG. Our outcomes indicate integrating MU discharges with modeled biomechanical outputs may provide a more accurate torque control signal than direct EMG control of assistive, lower-limb devices, such as exoskeletons and powered prostheses.