Biology subjects

Nazarpour, K.

Publications and source records attributed to Nazarpour, K..

13 recordsLinked to original sources

bioRxiv2025

Scalability of Random Forest in Myoelectric Control

ObjectiveMyoelectric control systems translate electromyographic (EMG) signals into control commands, enabling immersive human-robot interactions in the real world and the Metaverse. The variability of EMG due to various confounding factors leads to significant performance degradation. Such variability can be mitigated by training a highly generalisable but massively parameterized deep neural network, which can be effectively scaled using a vast dataset. We aim to find an alternative simple, explainable, efficient and parallelisable model, which can flexibly scale up with a larger dataset and scale down to reduce model size, will significantly facilitate the practical implementation of myoelectric control. ApproachIn this work, we discuss the scalability of a random forest (RF) for myoelectric control. We show how to scale an RF up and down during the process of pre-training, fine-tuning, and automatic self-calibration. The effects of diverse factors such as bootstrapping, decision tree editing (pre-training, pruning, grafting, appending), and the size of training data are systematically studied using EMG data from 106 participants including both low- and high-density electrodes. Main resultsWe examined several factors that affect the size and accuracy of the model. The best solution could reduce the size of RF models by{approx} 500 x, with the accuracy reduced by only 1.5%. Importantly, for the first time we report the unique merit of RF that with more EMG electrodes (higher input dimension), the RF model size would be reduced, contrasting all other models. SignificanceAll of these findings unlock the full potential of RF in real-world applications.

bioengineering↗

bioRxiv2024

An Open-Source Neurodynamic Model of the Bladder

BackgroundLower urinary tract (LUT) symptoms affect a significant proportion of the population. In-silico medicine can help understand these conditions and develop treatments. However, current LUT computational models are closed source, too deterministic, and do not allow for simple use of modelling neural intervention. MethodsAn open-source Python-based model was developed to simulate bladder, sphincter, and kidney dynamics using normalised neural signals to predict pressure and volume. The model was verified against animal bladder data, assessed for noise sensitivity, and evaluated against known physiological factors. ResultsThe animal data comparison yielded a significantly more similar pattern than existing models, with a correlation coefficient of r = 0.93 (p < 0.001). All physiological factors were within bounds, and the model remained stable with noise under the described boundaries. ConclusionThe proposed model advances the field of computational medicine by providing an open-source model for researchers and developers. It improves upon existing models by being accessible, including a built-in neural model that better replicates smooth bladder filling results, and incorporating a novel kidney function that alters bladder function by time of day in line with circadian rhythm. Future applications include personalised medicine, treating LUT symptoms with in-silico models and adaptive neural interventions.

bioinformatics↗

bioRxiv2024

Plug-and-Play Myoelectric Control via a Self-Calibrating Random Forest Common Model

ObjectiveElectromyographic (EMG) signals show large variabilities over time due to factors such as electrode shifting, user behaviour variations, etc., substantially degrading the performance of myoelectric control models in long-term use. Previously one-time model calibration was usually required each time before usage. However, the EMG characteristics could change even within a short period of time. Our objective is to develop a self-calibrating model, with an automatic and unsupervised self-calibration mechanism. ApproachWe developed a computationally efficient random forest (RF) common model, which can (1) be pre-trained and easily adapt to a new user via one-shot calibration, and (2) keep calibrating itself once in a while by boosting the RF with new decision trees trained on pseudo-labels of testing samples in a data buffer. Main resultsOur model has been validated in both offline and real-time, both open and closed-loop, both intra-day and long-term (up to 5 weeks) experiments. We tested this approach with data from 66 able-bodied participants. We also explored the effects of bidirectional user-model co-adaption in closed-loop experiments. We found that the self-calibrating model could gradually improve its performance in long-term use. With visual feedback, users will also adapt to the dynamic model meanwhile learn to perform hand gestures with significantly lower EMG amplitudes (less muscle effort). SignificanceOur random forest-approach provides a new alternative built on simple decision tree for myoelectric control, which is explainable, computationally efficient, and requires minimal data for model calibration. Source codes are avaiable at: https://github.com/MoveR-Digital-Health-and-Care-Hub/self-calibrating-rf

neuroscience↗

bioRxiv2024

Low Frequency Tibial Neuromodulation Increases Voiding Activity - A Computational Model

Despite widespread clinical adoption for disorders of incontinence such as overactive bladder, there remain unknowns surrounding the mechanism that underpins Tibial Nerve Stimulation (TNS). Current understanding suggests that TNS counteracts incontinence by the inhibition of brainstem and spinal cord activity. How this inhibition alters bladder function is not fully understood. We hypothesize that the supraspinal components of the system act as a high-pass filter, allowing voiding signals to proceed only when bladder filling reaches a critical level. Testing this hypothesis may explain how TNS is able to induce both an inhibitory and a little-explored excitatory effect on bladder activity in response to high-frequency (20 Hz) and low-frequency (1 Hz) stimulation, respectively. We performed a single-blinded trial in healthy human participants administered high and low-frequency Transcutaneous TNS. We also developed a computational model of the lower-urinary tract and control circuit to study the frequency-dependent effects of TNS. For the first time, we report a frequency-dependent effect of TNS via the ability to alter urge perception and up-regulate and down-regulate bladder activity, corroborating model predictions. These results provide a foundation for the development of targeted and effective TNS therapies, benefiting from in silico models. We hope that future clinical research will determine the efficacy of low-frequency TNS as a non-invasive treatment option for urinary retention. Significance StatementThis work describes an experimental study supported by a novel computational model that for the first time captures the frequency-dependent effects of tibial neuromodulation on the urinary control system in humans. The findings of the work provide critically important evidence for the role of the brainstem as a filter, which may explain the little-explored excitatory effect of tibial neuromodulation on bladder activity. These results have considerable clinical implications in the treatment of urinary retention, a condition for which there are at present very few non-invasive treatment options available.

neuroscience↗

bioRxiv2023

Source-Free Random Forest Model Calibration for Myoelectric Control

ObjectiveMost existing machine learning models for myoelectric control require a large amount of data to learn user-specific characteristics of the electromyographic (EMG) signals, which is burdensome. Our objective is to develop an approach to enable the calibration of a pre-trained model with minimal data from a new myoelectric user. ApproachWe trained a random forest model with EMG data from 20 people collected during the performance of multiple hand grips. To adapt the decision rules for a new user, first, the branches of the pre-trained decision trees were pruned using the validation data from the new user. Then new decision trees trained merely with data from the new user were appended to the pruned pre-trained model. ResultsReal-time myoelectric experiments with 18 participants over two days demonstrated the improved accuracy of the proposed approach when compared to benchmark user-specific random forest and the linear discriminant analysis models. Furthermore, the random forest model that was calibrated on day one for a new participant yielded significantly higher accuracy on day two, when compared to the benchmark approaches, which reflects the robustness of the proposed approach. SignificanceThe proposed model calibration procedure is completely source-free, that is, once the base model is pre-trained, no access to the source data from the original 20 people is required. Our work promotes the use of efficient, explainable, and simple models for myoelectric control.

neuroscience↗

bioRxiv2023

Reducing Motor Variability Enhances Myoelectric Control Robustness Across Limb Positions

BackgroundThe limb position effect is a multi-faceted problem, associated with decreased upper-limb prosthesis control acuity following a change in arm position. The many factors contributing to this problem can arise from distinct environmental or physiological sources. Despite their differences in origin, the effect of each factor manifests similarly as increased variability in the detected control signal. This variability can cause incorrect decoding of user intent, leading to dropped items or inability to use the prosthesis during activities of daily living. The general approach of previous research has attempted to limit the impact of the factors or better capture the variability with data abundance. In this paper we take an alternative approach and investigate the effect of reducing the variability of the control signal by improving the consistency of muscle activity with user training. MethodsParticipants underwent 4 days of myoelectric training with either concurrent or delayed feedback in a single arm position. During this time, they were trained to control a two-dimensional cursor using muscles in the forearm. At the end of training participants underwent a zero feedback retention test in multiple limb positions. In doing so, we tested how well the skill learned in a single limb position generalized to untrained positions. ResultsWe found that delayed feedback training led to more consistent muscle activity across both the trained and untrained limb positions. Analysis of patterns of activation in the delayed feedback group suggest a structured change in muscle activity occurs across arm positions. The structured changes allowed us to quantify the limb position effect by comparing trained to untrained arm positions. Different limb positions changed mean ECR and FCR muscle activity in the range of -4.3% to +18.7%. All participants were able to counter the limb position effect if given concurrent feedback, confirming our results align with existing findings. ConclusionsOur results demonstrate that myoelectric user-training can lead to the retention of motor skills that are more robust to limb position changes. This work highlights the importance of reducing motor variability with practice, prior to examining the underlying structure of muscle changes associated with limb position. These findings will be useful for the majority of myoelectric prosthesis control systems and will create better quality input data leading to more robust machine-learning based prosthesis control systems.

neuroscience↗

bioRxiv2020

3D-Printing and upper-limb prosthetic sockets; promises and pitfalls

Modernising the way upper-limb prosthetic sockets are made has seen limited progress. The casting techniques that are employed in clinics today resemble those developed over 50 years ago and there is still a heavy reliance on manual labour. Modern manufacturing methods such as 3D scanning and printing are often presented as ready-to-use solutions for producing low-cost functional devices, with public perceptions being largely shaped by the superficial media representation and advertising. The promise is that modern socket manufacturing methods can improve patient satisfaction, decrease manufacturing times and reduce the workload in the clinic. However, the perception in the clinical community is that total conversion to digital methods in a clinical environment is not straightforward. Anecdotally, there is currently a disconnect between those developing technology to produce prosthetic devices and the actual needs of clinicians and people with limb difference. In this paper, we demonstrate strengths and drawbacks of a fully digitised, low-cost trans-radial diagnostic socket making process, informed by clinical expertise. We present volunteer feedback on the digitally created sockets and provide expert commentary on the use of digital tools in upper-limb socket manufacturing. We show that it is possible to utilise 3D scanning and printing, but only if the process is informed by expert knowledge. We bring examples to demonstrate how and why the process may go wrong. Finally, we provide discussion on why progress in modernising the manufacturing of upper-limb sockets has been slow yet it is still too early to rule out digital methods.

bioengineering↗

bioRxiv2020

Discrete action control for prosthetic digits

ObjectiveWe aim to develop a paradigm for simultaneous and independent control of multiple degrees of freedom (DOFs) for upper-limb prostheses. ApproachWe introduce action control, a novel method to operate prosthetic digits with surface electromyography (EMG) based on multi-label, multi-class classification. At each time step, the decoder classifies movement intent for each controllable DOF into one of three categories: open, close, or stall (i.e., no movement). We implemented a real-time myoelectric control system using this method and evaluated it by running experiments with one unilateral and two bilateral amputees. Participants controlled a six-DOF bar interface on a computer display, with each DOF corresponding to a motor function available in multi-articulated prostheses. Main resultsWe show that action control can significantly and systematically outperform the state-of-the-art method of position control via multi-output regression in both task- and non-task-related measures. Improvements in median task performance ranged from 20.14% to 62.32% for individual participants. Analysis of a post-experimental survey revealed that all participants rated action higher than position control in a series of qualitative questions and expressed an overall preference for the former. SignificanceAction control has the potential to improve the dexterity of upper-limb prostheses. In comparison with regression-based systems, it only requires discrete instead of real-valued ground truth labels, typically collected with motion tracking systems. This feature makes the system both practical in a clinical setting and also suitable for bilateral amputation. This work is the first demonstration of myoelectric digit control in bilateral upper-limb amputees. Further investigation and pre-clinical evaluation are required to assess the translational potential of the method.

bioengineering↗

bioRxiv2020

Myoelectric digit action decoding with multi-label, multi-class classification: an offline analysis

The ultimate goal of machine learning-based myoelectric control is simultaneous and independent control of multiple degrees of freedom (DOFs), including wrist and digit artificial joints. For prosthetic finger control, regression-based methods are typically used to reconstruct position/velocity trajectories from surface electromyogram (EMG) signals. Although such methods have produced highly-accurate results in offline analyses, their success in real-time prosthesis control settings has been rather limited. In this work, we propose action decoding, a paradigm-shifting approach for independent, multi-digit movement intent decoding based on multi-label, multi-class classification. At each moment in time, our algorithm classifies movement action for each available DOF into one of three categories: open, close, or stall (i.e., no movement). Despite using a classifier as the decoder, arbitrary hand postures are possible with our approach. We analyse a public dataset previously recorded and published by us, comprising measurements from 10 able-bodied and two transradial amputee participants. We demonstrate the feasibility of using our proposed action decoding paradigm to predict movement action for all five digits as well as rotation of the thumb. We perform a systematic offline analysis by investigating the effect of various algorithmic parameters on decoding performance, such as feature selection and choice of classification algorithm and multi-output strategy. The outcomes of the offline analysis presented in this study will be used to inform the real-time implementation of our algorithm. In the future, we will further evaluate its efficacy with real-time control experiments involving upper-limb amputees.

bioengineering↗

bioRxiv2019

Model-based control of individual finger movements for prosthetic hand function

Prosthetic devices for hand difference have advanced considerably in recent years, to the point where the mechanical dexterity of a state-of-the-art prosthetic hand approaches that of the natural hand. Control options for users, however, have not kept pace, meaning that the new devices are not used to their full potential. Promising developments in control technology reported in the literature have met with limited commercial and clinical success. We have previously described a biomechanical model of the hand that could be used for prosthesis control. In this study, we report on three key elements of the biomechanical simulations relevant to prosthesis control: we show the performance of the model in replicating recorded hand kinematics and find average correlations of 0.89 between modelled and recorded motions; we show that the computational performance of the simulations is fast enough to achieve real-time control with a robotic hand in the loop; and we describe the use of the model for controlling object gripping. Despite some limitations in accessing sufficient driving signals, the model performance shows promise as a controller for prosthetic hands when driven with recorded EMG signals. We identify areas for future work to address these limitations.

bioengineering↗

bioRxiv2019

Effect of motor learning on prosthetic finger controlwith an intuitive myoelectric decoder

Machine learning-based myoelectric control is regarded as an intuitive paradigm, because of the mapping it creates between muscle co-activation patterns and prosthesis movements that aims to simulate the physiological pathways found in the human arm. Despite that, there has been evidence that closed-loop interaction with a classification-based interface results in user adaptation, which leads to performance improvement with experience. Recently, there has been a focus shift towards continuous prosthesis control, yet little is known about whether and how user adaptation affects myoelectric control performance in dexterous, intuitive tasks. We investigate the effect of short-term adaptation with independent finger position control by conducting real-time experiments with 10 able-bodied and two transradial amputee subjects. We demonstrate that despite using an intuitive decoder, experience leads to significant improvements in performance. We argue that this is due to the lack of an utterly natural control scheme, which is mainly caused by differences in the anatomy of human and artificial hands, movement intent decoding inaccuracies, and lack of proprioception. Finally, we extend previous work in classification-based and wrist continuous control by verifying that offline analyses cannot reliably predict real-time performance, thereby reiterating the importance of validating myoelectric control algorithms with real-time experiments.

bioengineering↗

bioRxiv2019

Multi-grip classification-based prosthesis control with two sensors

In the field of upper-limb myoelectric prosthesis control, the use of statistical and machine learning methods has been long proposed as a means of enabling intuitive grip selection and actuation. Recently, this paradigm has found its way toward commercial adoption. Machine learning-based prosthesis control typically relies on the use of a large number of electrodes. Here, we propose an end-to-end strategy for multi-grip, classification-based prosthesis control using only two sensors, comprising electromyography (EMG) electrodes and inertial measurement units (IMUs). We emphasize the importance of accurately estimating posterior class probabilities and rejecting predictions made with low confidence, so as to minimize the rate of unintended prosthesis activations. To that end, we propose a confidence-based error rejection strategy using grip-specific thresholds. We evaluate the efficacy of the proposed system with real-time pick and place experiments using a commercial multi-articulated prosthetic hand and involving 12 able-bodied and two transradial (i.e., below-elbow) amputee participants. Results promise the potential for deploying intuitive, classification-based multi-grip control in existing upper-limb prosthetic systems subject to small modifications.

bioengineering↗

bioRxiv2017

Recovery Dynamics of the High Frequency Alternating Current Nerve Block

ObjectiveHigh-Frequency alternating current (HFAC) nerve block has great potential for neuromodulation-based therapies. However nerve function recovery dynamics after a block is highly understudied. This study aims to characterise the recovery dynamics of neural function after an HFAC block.\n\nApproachExperiments were carried out in-vivo to determine blocking efficacy as a function of blocking signal amplitude and frequency, and recovery times as well as recovery completeness was measured within a 0.7 s time scale from the end of block. The sciatic nerve was stimulated at 100 Hz during recovery to reduce error to within {+/-}10 ms for measurements of recovery dynamics. The electromyogram (EMG) signals were measured from gastrocnemius medialis and tibialis anterior during trials as an indicator for nerve function.\n\nMain ResultsThe HFAC block was most reliable around 20 kHz, with block thresholds approximately 5 or 6 mA depending on the animal and muscle. Recovery times ranged from 20 to 430 milliseconds and final values spanned relative outputs from approximately 1 to 0.2. Higher blocking signal frequencies and amplitudes increased recovery time and decreased recovery completeness.\n\nSignificanceThese results confirm that recovery dynamics from block depend on blocking signal frequency and amplitude, which is of particular importance for neuromodulation therapies and for comparing results across studies using different blocking signal parameters.

neuroscience↗