pith. sign in

arxiv: 2506.16468 · v2 · submitted 2025-06-19 · 💻 cs.HC · cs.SY· eess.SY

Closed-loop Neuroprosthetic Control through Spared Neural Activity Enables Proportional Foot Movements after Spinal Cord Injury

Vlad Cnejevici , Matthias Ponfick , Dietmar Fey , Raul C. S\^impetru , Alessandro Del Vecchio This is my paper

Pith reviewed 2026-05-19 08:27 UTC · model grok-4.3

classification 💻 cs.HC cs.SYeess.SY
keywords spinal cord injuryelectromyographyfunctional electrical stimulationneuroprostheticsclosed-loop controlproportional controlfoot movementwearable device
0
0 comments X

The pith

Decoded EMG signals from spared neural activity after spinal cord injury let users voluntarily control functional electrical stimulation for proportional foot movements.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper shows that surface EMG recorded from the lower limb below a spinal cord injury can be decoded in real time to drive a functional electrical stimulation device in closed loop. This setup lets participants generate voluntary foot flexion, extension, and side-to-side movements instead of relying on motion sensors or open-loop triggers. Two users increased active foot flexion range to 33.6 percent and 40 percent of a typical healthy functional range, while one user selected among up to six distinct stimulation intensities with good accuracy. The approach uses only a wearable 32-channel bracelet and standard skin electrodes, offering a non-invasive route to restore lower-limb function that many people with SCI lose.

Core claim

A wearable system records 32-channel EMG from the affected leg, applies machine learning to decode spared movement intent, and feeds the resulting control signal directly into an FES device. In closed-loop use, two participants produced significantly larger foot flexion excursions than without the decoded command, reaching 33.6 percent and 40 percent of healthy range. One participant further demonstrated voluntary proportional control by selecting among as many as six stimulation levels during flexion and extension tasks.

What carries the argument

The real-time machine-learning decoder that converts multi-channel EMG patterns into proportional stimulation commands for the FES device.

If this is right

  • Users can produce distinct activation patterns for foot flexion, extension, and inversion or eversion on command.
  • Proportional control of two or more stimulation levels is achievable with accuracy above 70 percent.
  • Closed-loop stimulation driven by decoded EMG yields statistically reliable increases in active foot range of motion.
  • The same signals support voluntary rather than purely reactive triggering of stimulation.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • A wearable bracelet form factor may allow the approach to move from lab sessions to daily home or community use.
  • The decoding strategy could be tested on other joints or combined with gait training to support walking.
  • If the spared EMG patterns remain stable over months, the method might reduce the need for compensatory trunk or hip strategies that often cause secondary pain.

Load-bearing premise

The EMG activity captured below the injury must reflect genuine voluntary motor intent from spared neural pathways rather than compensatory movements, fatigue, or electrical cross-talk.

What would settle it

A trial in which the same participants perform the foot tasks while deliberately suppressing voluntary intent and the system is still allowed to stimulate based on any residual EMG would show whether the measured range gains disappear.

Figures

Figures reproduced from arXiv: 2506.16468 by Alessandro Del Vecchio, Dietmar Fey, Matthias Ponfick, Raul C. S\^impetru, Vlad Cnejevici.

Figure 1
Figure 1. Figure 1: Study overview. (A) Five individuals with SCI ( [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Experimental setup. (A) Participants attempted ankle movements while spared neural activity was recorded using a 32- channel EMG bracelet and wirelessly streamed to a host device. For the electrical stimulation test, their ankle RoM before and while delivering stimulation pulses was captured using a motion capture camera system and infrared reflective markers placed on the shin and foot. (B) A 2D reference… view at source ↗
Figure 3
Figure 3. Figure 3: Overview of motor dimension separation test. (A) All participants attempted different ankle movements while following a reference cursor (coordinates normalized between -1 and +1) along the X-axis (for inversion/eversion) or Y-axis (for dorsi- /plantarflexion). Based on the ML model prediction, the position of a second cursor was updated incrementally (see Materials and Methods). During post-processing, X-… view at source ↗
Figure 4
Figure 4. Figure 4: Example of MU activity during foot dorsi-/plantarflexion. (A) Raw EMG data was recorded during foot dorsi- /plantarflexion (here shown for S2, [PITH_FULL_IMAGE:figures/full_fig_p010_4.png] view at source ↗
read the original abstract

Loss of voluntary foot movement after spinal cord injury (SCI) can significantly limit independent mobility and quality of life. To improve motor output after injury, functional electrical stimulation (FES) is used to deliver stimulation pulses through the skin to affected muscles. While commercial FES systems typically use motion-based triggers, prior research shows that spared movement intent can be decoded after SCI using surface electromyography (EMG). Our aim is to assess how well spared neural signals of the lower limb after SCI can be decoded and used to control electrical stimulation for restoring foot movement. We developed a wearable machine learning-powered neuroprosthetic that records EMG from the affected lower limb using a 32-channel electrode bracelet and enables closed-loop control of a FES device for foot movement restoration. Five participants with SCI used the predicted control signal to follow trajectories on a screen with their foot and achieve distinct motor activation patterns for foot flexion, extension, and inversion or eversion. Three of these participants also achieved 2 proportional activation levels during foot flexion/extension with more than 70% accuracy. To validate how these neural signals can be used for closed-loop neuroprosthetic control, two participants used their decoded activity to control a FES device and stimulate their affected foot. This resulted in an increased foot flexion range for both participants of 33.6% and 40% of a functional healthy range, respectively (p smaller than 0.001). One of the participants also achieved voluntary proportional control of up to 6 stimulation levels during foot flexion/extension. These results suggest that wearable EMG decoding coupled with FES systems provides a scalable strategy for closed-loop neuroprosthetic control supporting voluntary foot movement.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The paper claims that a wearable 32-channel EMG bracelet combined with machine learning can decode spared lower-limb motor intent after SCI and drive closed-loop FES to restore proportional foot flexion/extension. In a pilot with five participants, distinct activation patterns were decoded; two participants then used the decoded signals for closed-loop FES, achieving foot-flexion range increases of 33.6 % and 40 % of a functional healthy range (p < 0.001), with one participant reaching voluntary proportional control across up to six stimulation levels.

Significance. If the decoded EMG truly indexes spared voluntary drive rather than compensatory or artifactual activity, the work demonstrates a non-invasive, wearable route to closed-loop neuroprosthetic control that could meaningfully improve mobility for individuals with SCI. The proportional, multi-level control result is functionally relevant and the wearable form factor is scalable.

major comments (2)
  1. [Abstract and Results] Abstract and Results (participant performance paragraphs): The central claim that the decoded 32-channel EMG represents genuine spared voluntary motor intent from below the lesion is load-bearing for interpreting the closed-loop FES outcomes as restoration of voluntary function. The manuscript provides no quantitative controls (e.g., simultaneous proximal-muscle recordings, instructed compensatory-strategy trials, or fatigue protocols) to disambiguate volume conduction, cross-talk, or non-specific activation from true below-lesion drive.
  2. [Methods and Results] Methods and Results: Model training details, cross-validation procedure, electrode-placement standardization, and explicit criteria for excluding movement artifacts are not reported. Given the small sample (n=5 for decoding, n=2 for closed-loop), these omissions make it difficult to assess whether the reported >70 % accuracy for two proportional levels and the statistically significant range increases are robust.
minor comments (2)
  1. [Abstract] Abstract: 'p smaller than 0.001' should be written as the conventional 'p < 0.001'.
  2. [Methods] Figure legends and Methods: Clarify the exact feature set and classifier architecture used for the 32-channel EMG decoding to allow replication.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We appreciate the referee's insightful comments on our manuscript. Below we provide point-by-point responses to the major comments and indicate the revisions we will make.

read point-by-point responses

  1. Referee: [Abstract and Results] Abstract and Results (participant performance paragraphs): The central claim that the decoded 32-channel EMG represents genuine spared voluntary motor intent from below the lesion is load-bearing for interpreting the closed-loop FES outcomes as restoration of voluntary function. The manuscript provides no quantitative controls (e.g., simultaneous proximal-muscle recordings, instructed compensatory-strategy trials, or fatigue protocols) to disambiguate volume conduction, cross-talk, or non-specific activation from true below-lesion drive.

    Authors: We agree that additional quantitative controls would strengthen the claim that the decoded signals index spared voluntary drive rather than volume conduction or compensatory activity. The pilot study did not include simultaneous proximal-muscle recordings, dedicated compensatory-strategy trials, or fatigue protocols. We will revise the manuscript to add a dedicated limitations paragraph in the Discussion that explicitly acknowledges this gap, describes the instructions given to participants to perform isolated foot movements, and outlines how future work could incorporate the suggested controls. revision: yes

  2. Referee: [Methods and Results] Methods and Results: Model training details, cross-validation procedure, electrode-placement standardization, and explicit criteria for excluding movement artifacts are not reported. Given the small sample (n=5 for decoding, n=2 for closed-loop), these omissions make it difficult to assess whether the reported >70 % accuracy for two proportional levels and the statistically significant range increases are robust.

    Authors: We thank the referee for highlighting these reporting omissions. In the revised manuscript we will expand the Methods section to detail the machine learning model architecture and training procedure, the cross-validation scheme used, the anatomical landmarks and protocol for standardizing electrode-bracelet placement, and the explicit criteria (signal amplitude thresholds and visual inspection) applied to exclude movement artifacts. We will also reframe the Results to emphasize the pilot nature of the work (n=5 and n=2) and include per-participant performance metrics so readers can evaluate robustness directly. revision: yes

Circularity Check

0 steps flagged

Empirical demonstration with no derivation chain

full rationale

This is a pilot empirical study reporting measured performance outcomes from five SCI participants using a wearable EMG-decoding system for closed-loop FES control. The key results (33.6% and 40% of healthy foot flexion range, p<0.001; up to 6 proportional stimulation levels) are direct experimental observations from participant trials, not outputs of any mathematical derivation, fitted parameter renamed as prediction, or self-referential definition. No equations, uniqueness theorems, or ansatzes appear in the provided text that would reduce the reported metrics to the input data by construction. The machine-learning decoding step is a standard applied technique whose accuracy is evaluated against the same participants' voluntary attempts, but the final range-of-motion and accuracy figures remain independent empirical measurements rather than tautological restatements. Minor self-citation risk is absent from the abstract and described methods.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The central claim rests on standard assumptions in EMG signal processing and FES safety rather than new postulates. No free parameters are explicitly fitted in the abstract; the ML decoder parameters are implicit but not enumerated. No new entities are invented.

axioms (2)
  • domain assumption Surface EMG from the affected limb after SCI contains decodable voluntary motor intent that can be distinguished from noise and compensatory activity.
    Invoked in the abstract when stating that 'spared movement intent can be decoded' and used for closed-loop control.
  • domain assumption Functional electrical stimulation delivered via skin electrodes can produce graded, functional foot movements without causing discomfort or fatigue that would invalidate voluntary control.
    Required for the reported range increases and proportional stimulation levels to be interpreted as restored voluntary movement.

pith-pipeline@v0.9.0 · 5864 in / 1770 out tokens · 30578 ms · 2026-05-19T08:27:49.677718+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

54 extracted references · 54 canonical work pages · 1 internal anchor

  1. [1]

    Gait Impairments in a Group of Patients With Incomplete Spinal Cord Injury and Their Relevance Regarding Therapeutic Approaches Using Functional Electrical Stim ulation,

    A. Van Der Salm, A. V. Nene, D. J. Maxwell, P. H. Veltink, H. J. Hermens, and M. J. IJzerman, “Gait Impairments in a Group of Patients With Incomplete Spinal Cord Injury and Their Relevance Regarding Therapeutic Approaches Using Functional Electrical Stim ulation,” Artificial Organs , vol. 29, no. 1, pp. 8 –14, Jan. 2005, doi: 10.1111/j.1525 - 1594.2004.29004.x

    work page doi:10.1111/j.1525 2005

  2. [2]

    Behandlungsoptionen beim neurogenen Lähmungsfuß – eine systematische Literaturrecherche,

    E. Jakubowitz, D. Yao, H. Windhagen, C. Stukenborg -Colsman, A. Thomann, and K. Daniilidis, “Behandlungsoptionen beim neurogenen Lähmungsfuß – eine systematische Literaturrecherche,” Z Orthop Unfall, vol. 155, no. 04, pp. 402 –408, Aug. 2017, doi: 10.1055/s-0043-100760

    work page doi:10.1055/s-0043-100760 2017

  3. [3]

    Multicenter study of peroneal mononeuropathy: clinical, neurophysiologic, and quality of life assessment,

    I. Aprile et al., “Multicenter study of peroneal mononeuropathy: clinical, neurophysiologic, and quality of life assessment,” J Peripheral Nervous Sys, vol. 10, no. 3, pp. 259–268, Sep. 2005, doi: 10.1111/j.1085-9489.2005.10304.x

    work page doi:10.1111/j.1085-9489.2005.10304.x 2005

  4. [4]

    Neuroprosthetics: from sensorimotor to cognitive disorders,

    A. Gupta, N. Vardalakis, and F. B. Wagner, “Neuroprosthetics: from sensorimotor to cognitive disorders,” Commun Biol, vol. 6, no. 1, p. 14, Jan. 2023, doi: 10.1038/s42003-022-04390-w

    work page doi:10.1038/s42003-022-04390-w 2023

  5. [5]

    Effects of a New Radio Frequency –Controlled Neuroprosthesis on Gait Symmetry and Rhythmicity in Patients with Chronic Hemiparesis,

    J. M. Hausdorff and H. Ring, “Effects of a New Radio Frequency –Controlled Neuroprosthesis on Gait Symmetry and Rhythmicity in Patients with Chronic Hemiparesis,” American Journal of Physical Medicine & Rehabilitation, vol. 87, no. 1, pp. 4–13, Jan. 2008, doi: 10.1097/PHM.0b013e31815e6680

    work page doi:10.1097/phm.0b013e31815e6680 2008

  6. [6]

    Quantification of gait kinematics and walking ability of people with multiple sclerosis who are new users of functional electrical stimulation,

    S. Scott, M. Linden, J. Hooper, P. Cowan, and T. Mercer, “Quantification of gait kinematics and walking ability of people with multiple sclerosis who are new users of functional electrical stimulation,” J Rehabil Med, vol. 45, no. 4, pp. 364–369, 2013, doi: 10.2340/16501977-1109

    work page doi:10.2340/16501977-1109 2013

  7. [7]

    Kinematic and kinetic benefits of implantable peroneal nerve stimulation in people with post -stroke drop foot using an ankle -foot orthosis,

    F. Berenpas, S. Schiemanck, A. Beelen, F. Nollet, V. Weerdesteyn, and A. Geurts, “Kinematic and kinetic benefits of implantable peroneal nerve stimulation in people with post -stroke drop foot using an ankle -foot orthosis,” Restorative Neurology and Neuroscience, vol. 36, no. 4, pp. 547–558, Jul. 2018, doi: 10.3233/RNN-180822

    work page doi:10.3233/rnn-180822 2018

  8. [8]

    A decision support system for electrode shaping in multi -pad FES foot drop correction,

    J. Malešević et al. , “A decision support system for electrode shaping in multi -pad FES foot drop correction,” J NeuroEngineering Rehabil, vol. 14, no. 1, p. 66, Dec. 2017, doi: 10.1186/s12984-017-0275-5

    work page doi:10.1186/s12984-017-0275-5 2017

  9. [9]

    Automated setup of functional electrical stimulation for drop foot using a novel 64 channel prototype stimulator and electrode array: Results from a gait-lab based study,

    B. W. Heller et al., “Automated setup of functional electrical stimulation for drop foot using a novel 64 channel prototype stimulator and electrode array: Results from a gait-lab based study,” Medical Engineering & Physics, vol. 35, no. 1, pp. 74– 81, Jan. 2013, doi: 10.1016/j.medengphy.2012.03.012

    work page doi:10.1016/j.medengphy.2012.03.012 2013

  10. [10]

    Functional Electrical Stimulation for Foot Drop Injury Based on the Arm Swing Motion,

    S. Ismail, M. N. Harun, and A. H. Omar, “Functional Electrical Stimulation for Foot Drop Injury Based on the Arm Swing Motion,” Procedia Manufacturing, vol. 2, pp. 490–494, 2015, doi: 10.1016/j.promfg.2015.07.084

    work page doi:10.1016/j.promfg.2015.07.084 2015

  11. [11]

    A functional electrical stimulation system for human walking inspired by reflexive control principles,

    L. Meng, B. Porr, C. A. Macleod, and H. Gollee, “A functional electrical stimulation system for human walking inspired by reflexive control principles,” Proc Inst Mech Eng H , vol. 231, no. 4, pp. 315 –325, Apr. 2017, doi: 10.1177/0954411917693879

    work page doi:10.1177/0954411917693879 2017

  12. [12]

    A novel motion sensor -driven control system for FES -assisted walking after spinal cord injury: A pilot study,

    G. P. Braz, M. F. Russold, C. Fornusek, N. A. Hamzaid, R. M. Smith, and G. M. Davis, “A novel motion sensor -driven control system for FES -assisted walking after spinal cord injury: A pilot study,” Medical Engineering & Physics, vol. 38, no. 11, pp. 1223–1231, Nov. 2016, doi: 10.1016/j.medengphy.2016.06.007

    work page doi:10.1016/j.medengphy.2016.06.007 2016

  13. [13]

    Closed‐Loop Wearable Device Network of Intrinsically‐Controlled, Bilateral Coordinated Functional Electrical Stimulation for Stroke,

    S. Xu et al. , “Closed‐Loop Wearable Device Network of Intrinsically‐Controlled, Bilateral Coordinated Functional Electrical Stimulation for Stroke,” Advanced Science, vol. 11, no. 17, p. 2304763, May 2024, doi: 10.1002/advs.202304763

    work page doi:10.1002/advs.202304763 2024

  14. [14]

    Machine -learning-based coordination of powered ankle –foot orthosis and functional electrical stimulation for gait control,

    S. Jung, J. H. Bong, K. Kim, and S. Park, “Machine -learning-based coordination of powered ankle –foot orthosis and functional electrical stimulation for gait control,” Front. Bioeng. Biotechnol. , vol. 11, p. 1272693, Jan. 2024, doi: 10.3389/fbioe.2023.1272693

    work page doi:10.3389/fbioe.2023.1272693 2024

  15. [15]

    An EMG-triggered cooperative controller for a hybrid FES-robotic system,

    F. Ferrari, E. Zimei, M. Gandolla, A. Pedrocchi, and E. Ambrosini, “An EMG-triggered cooperative controller for a hybrid FES-robotic system,” in 2023 IEEE International Conference on Metrology for eXtended Reality, Artificial Intelligence and Neural Engineering (MetroXRAINE) , Milano, Italy: IEEE, Oct. 2023, pp. 852 –857. doi: 10.1109/MetroXRAINE58569.202...

    work page doi:10.1109/metroxraine58569.2023.10405687 2023

  16. [16]

    A Novel sEMG Triggered FES-Hybrid Robotic Lower Limb Rehabilitation System for Stroke Patients,

    I. L. Petersen, W. Nowakowska, C. Ulrich, and L. N. S. A. Struijk, “A Novel sEMG Triggered FES-Hybrid Robotic Lower Limb Rehabilitation System for Stroke Patients,” IEEE Trans. Med. Robot. Bionics, vol. 2, no. 4, pp. 631–638, Nov. 2020, doi: 10.1109/TMRB.2020.3019081

    work page doi:10.1109/tmrb.2020.3019081 2020

  17. [17]

    Merletti and P

    R. Merletti and P. Parker, Eds., Electromyography: Physiology, Engineering, and Noninvasive Applications, 1st ed. Wiley,

    work page

  18. [18]

    doi: 10.1002/0471678384

    work page doi:10.1002/0471678384

  19. [19]

    Comparison of stress detection through ecg and ppg signals using a random forest-based algorithm,

    A. L. Cakici et al., “A Generalized Framework for the Study of Spinal Motor Neurons Controlling the Human Hand During Dynamic Movements,” in 2022 44th Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC) , Glasgow, Scotland, United Kingdom: IEEE, Jul. 2022, pp. 4115 –4118. doi: 10.1109/EMBC48229.2022.9870914

    work page doi:10.1109/embc48229.2022.9870914 2022

  20. [20]

    A direct spinal cord–computer interface enables the control of the paralysed hand in spinal cord injury,

    D. S. Oliveira et al., “A direct spinal cord–computer interface enables the control of the paralysed hand in spinal cord injury,” Brain, vol. 147, no. 10, pp. 3583–3595, Oct. 2024, doi: 10.1093/brain/awae088

    work page doi:10.1093/brain/awae088 2024

  21. [21]

    MyoGestic: EMG interfacing framework for decoding multiple spared motor dimensions in individuals with neural lesions,

    R. C. Sîmpetru et al. , “MyoGestic: EMG interfacing framework for decoding multiple spared motor dimensions in individuals with neural lesions,” Sci. Adv., vol. 11, no. 15, p. eads9150, Apr. 2025, doi: 10.1126/sciadv.ads9150

    work page doi:10.1126/sciadv.ads9150 2025

  22. [22]

    Simultaneous control of multiple functions of bionic hand prostheses: Performance and robustness in end users,

    J. M. Hahne, M. A. Schweisfurth, M. Koppe, and D. Farina, “Simultaneous control of multiple functions of bionic hand prostheses: Performance and robustness in end users,” Sci. Robot. , vol. 3, no. 19, p. eaat3630, Jun. 2018, doi: 10.1126/scirobotics.aat3630

    work page doi:10.1126/scirobotics.aat3630 2018

  23. [23]

    Affordable Embroidered EMG Electrodes for Myoelectric Control of Prostheses: A Pilot Study,

    E. N. Kamavuako, M. Brown, X. Bao, I. Chihi, S. Pitou, and M. Howard, “Affordable Embroidered EMG Electrodes for Myoelectric Control of Prostheses: A Pilot Study,” Sensors, vol. 21, no. 15, p. 5245, Aug. 2021, doi: 10.3390/s21155245

    work page doi:10.3390/s21155245 2021

  24. [24]

    Simultaneous and Proportional Real-Time Myocontrol of Up to Three Degrees of Freedom of the Wrist and Hand,

    M. Nowak, I. Vujaklija, A. Sturma, C. Castellini, and D. Farina, “Simultaneous and Proportional Real-Time Myocontrol of Up to Three Degrees of Freedom of the Wrist and Hand,” IEEE Trans. Biomed. Eng., vol. 70, no. 2, pp. 459–469, Feb. 2023, doi: 10.1109/TBME.2022.3194104

    work page doi:10.1109/tbme.2022.3194104 2023

  25. [25]

    International Standards for Neurological Classification of Spinal Cord Injury,

    R. Rupp et al., “International Standards for Neurological Classification of Spinal Cord Injury,” Topics in Spinal Cord Injury Rehabilitation, vol. 27, no. 2, pp. 1–22, Mar. 2021, doi: 10.46292/sci2702-1

    work page doi:10.46292/sci2702-1 2021

  26. [26]

    Muscle Strength Grading,

    U. Naqvi and A. L. Sherman, “Muscle Strength Grading,” in StatPearls, Treasure Island (FL): StatPearls Publishing, 2025. Accessed: Apr. 24, 2025. [Online]. Available: http://www.ncbi.nlm.nih.gov/books/NBK436008/

    work page 2025

  27. [27]

    Scikit-learn: Machine Learning in Python,

    F. Pedregosa et al., “Scikit-learn: Machine Learning in Python,” MACHINE LEARNING IN PYTHON

    work page

  28. [28]

    CatBoost: unbiased boosting with categorical features

    L. Prokhorenkova, G. Gusev, A. Vorobev, A. V. Dorogush, and A. Gulin, “CatBoost: unbiased boosting with categorical features”

    work page

  29. [29]

    Classification of ankle joint movements based on surface electromyography signals for rehabilitation robot applications,

    M. S. AL -Quraishi et al. , “Classification of ankle joint movements based on surface electromyography signals for rehabilitation robot applications,” Med Biol Eng Comput, vol. 55, no. 5, pp. 747–758, May 2017, doi: 10.1007/s11517-016- 1551-4

    work page doi:10.1007/s11517-016- 2017

  30. [30]

    Decoding of Ankle Joint Movements in Stroke Patients Using Surface Electromyography,

    A. Noor et al., “Decoding of Ankle Joint Movements in Stroke Patients Using Surface Electromyography,” Sensors, vol. 21, no. 5, p. 1575, Feb. 2021, doi: 10.3390/s21051575

    work page doi:10.3390/s21051575 2021

  31. [31]

    A case study on classification of foot gestures via surface electromyography,

    K. R. Lyons and S. S. Joshi, “A case study on classification of foot gestures via surface electromyography,” presented at the Proc. Annu. Conf. Rehabil. Eng. Assist. Technol. Soc. Amer, 2015, pp. 1–5

    work page 2015

  32. [32]

    Akiba, S

    T. Akiba, S. Sano, T. Yanase, T. Ohta, and M. Koyama, “Optuna: A Next -generation Hyperparameter Optimization Framework,” in Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, Anchorage AK USA: ACM, Jul. 2019, pp. 2623–2631. doi: 10.1145/3292500.3330701

    work page doi:10.1145/3292500.3330701 2019

  33. [33]

    Procrustes Methods in the Statistical Analysis of Shape,

    C. Goodall, “Procrustes Methods in the Statistical Analysis of Shape,” Journal of the Royal Statistical Society Series B: Statistical Methodology, vol. 53, no. 2, pp. 285–321, Jan. 1991, doi: 10.1111/j.2517-6161.1991.tb01825.x

    work page doi:10.1111/j.2517-6161.1991.tb01825.x 1991

  34. [34]

    Fluctuations in isometric muscle force can be described by one linear projection of low‐frequency components of motor unit discharge rates,

    F. Negro, A. Holobar, and D. Farina, “Fluctuations in isometric muscle force can be described by one linear projection of low‐frequency components of motor unit discharge rates,” The Journal of Physiology, vol. 587, no. 24, pp. 5925–5938, Dec. 2009, doi: 10.1113/jphysiol.2009.178509

    work page doi:10.1113/jphysiol.2009.178509 2009

  35. [35]

    Divergence Measures Based on the Shannon Entropy

    J. Lin, “Divergence measures based on the Shannon entropy,” IEEE Trans. Inform. Theory, vol. 37, no. 1, pp. 145–151, Jan. 1991, doi: 10.1109/18.61115

    work page doi:10.1109/18.61115 1991

  36. [36]

    Spasticity-assessment: a review,

    F. Biering-Sørensen, J. B. Nielsen, and K. Klinge, “Spasticity-assessment: a review,” Spinal Cord, vol. 44, no. 12, pp. 708– 722, Dec. 2006, doi: 10.1038/sj.sc.3101928

    work page doi:10.1038/sj.sc.3101928 2006

  37. [37]

    Active and passive contributions to joint kinetics during walking in older adults,

    A. Silder, B. Heiderscheit, and D. G. Thelen, “Active and passive contributions to joint kinetics during walking in older adults,” Journal of Biomechanics, vol. 41, no. 7, pp. 1520–1527, Jan. 2008, doi: 10.1016/j.jbiomech.2008.02.016

    work page doi:10.1016/j.jbiomech.2008.02.016 2008

  38. [38]

    Estimating gait parameters from sEMG signals using machine learning techniques under different power capacity of muscle,

    S.-H. Liu, A. K. Sharma, B.-Y. Wu, X. Zhu, C.-J. Chang, and J.-J. Wang, “Estimating gait parameters from sEMG signals using machine learning techniques under different power capacity of muscle,” Sci Rep, vol. 15, no. 1, p. 12575, Apr. 2025, doi: 10.1038/s41598-025-95973-0

    work page doi:10.1038/s41598-025-95973-0 2025

  39. [39]

    Classification of Spinal Cord Injured EMG Data for Locomotion Recovery,

    R. Q. Y. Chia, “Classification of Spinal Cord Injured EMG Data for Locomotion Recovery,” University of Technology Sydney (Australia), 2024

    work page 2024

  40. [40]

    The discharge characteristics of motor units innervating functionally paralyzed muscles,

    D. S. D. Oliveira, M. Carbonaro, B. J. Raiteri, A. Botter, M. Ponfick, and A. Del Vecchio, “The discharge characteristics of motor units innervating functionally paralyzed muscles,” Journal of Neurophysiology , vol. 133, no. 2, pp. 343 –357, Feb. 2025, doi: 10.1152/jn.00389.2024

    work page doi:10.1152/jn.00389.2024 2025

  41. [41]

    Observation of inclusive $D^{*\pm}$ production in the decay of $\Upsilon(1S)$

    M. Oßwald, A. L. Cakici, D. Souza De Oliveira, D. I. Braun, D. Farina, and A. Del Vecchio, “Task -specific motor units in the extrinsic hand muscles control single- and multidigit tasks of the human hand,” Journal of Applied Physiology, vol. 138, no. 5, pp. 1187–1200, May 2025, doi: 10.1152/japplphysiol.00911.2024

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1152/japplphysiol.00911.2024 2025

  42. [42]

    The effect of subtalar inversion/eversion on the dynamic function of the tibialis anterior, soleus, and gastrocnemius during the stance phase of gait,

    R. Wang and E. M. Gutierrez -Farewik, “The effect of subtalar inversion/eversion on the dynamic function of the tibialis anterior, soleus, and gastrocnemius during the stance phase of gait,” Gait & Posture, vol. 34, no. 1, pp. 29 –35, May 2011, doi: 10.1016/j.gaitpost.2011.03.003. 26

    work page doi:10.1016/j.gaitpost.2011.03.003 2011

  43. [43]

    (2024, Mar 05)

    A. M. Van Leeuwen, J. H. Van Dieën, A. Daffertshofer, and S. M. Bruijn, “Ankle muscles drive mediolateral center of pressure control to ensure stable steady state gait,” Sci Rep, vol. 11, no. 1, p. 21481, Nov. 2021, doi: 10.1038/s41598 -021- 00463-8

    work page doi:10.1038/s41598 2021

  44. [44]

    Characterizing Natural Recovery after Traumatic Spinal Cord Injury,

    S. Kirshblum, B. Snider, F. Eren, and J. Guest, “Characterizing Natural Recovery after Traumatic Spinal Cord Injury,” Journal of Neurotrauma, vol. 38, no. 9, pp. 1267–1284, May 2021, doi: 10.1089/neu.2020.7473

    work page doi:10.1089/neu.2020.7473 2021

  45. [45]

    Resistance training and locomotor recovery after incomplete spinal cord injury: a case series,

    C. M. Gregory et al., “Resistance training and locomotor recovery after incomplete spinal cord injury: a case series,” Spinal Cord, vol. 45, no. 7, pp. 522–530, Jul. 2007, doi: 10.1038/sj.sc.3102002

    work page doi:10.1038/sj.sc.3102002 2007

  46. [46]

    Identification of Spared and Proportionally Controllable Hand Motor Dimensions in Motor Complete Spinal Cord Injuries Using Latent Manifold Analysis,

    R. C. Sîmpetru, D. Souza De Oliveira, M. Ponfick, and A. Del Vecchio, “Identification of Spared and Proportionally Controllable Hand Motor Dimensions in Motor Complete Spinal Cord Injuries Using Latent Manifold Analysis,” IEEE Trans. Neural Syst. Rehabil. Eng., vol. 32, pp. 3741–3750, 2024, doi: 10.1109/TNSRE.2024.3472063

    work page doi:10.1109/tnsre.2024.3472063 2024

  47. [47]

    Rear-foot, mid-foot and fore-foot motion during the stance phase of gait,

    A. Leardini, M. G. Benedetti, L. Berti, D. Bettinelli, R. Nativo, and S. Giannini, “Rear-foot, mid-foot and fore-foot motion during the stance phase of gait,” Gait & Posture, vol. 25, no. 3, pp. 453–462, Mar. 2007, doi: 10.1016/j.gaitpost.2006.05.017

    work page doi:10.1016/j.gaitpost.2006.05.017 2007

  48. [48]

    Ankle joint dorsiflexion. Establishment of a normal range,

    B. Baggett and G. Young, “Ankle joint dorsiflexion. Establishment of a normal range,” J Am Podiatr Med Assoc, vol. 83, no. 5, pp. 251–254, May 1993, doi: 10.7547/87507315-83-5-251

    work page doi:10.7547/87507315-83-5-251 1993

  49. [49]

    Sweigart, asweigart/pyautogui

    A. Sweigart, asweigart/pyautogui. (Jan. 14, 2025). Python. Accessed: Jan. 14, 2025. [Online]. Available: https://github.com/asweigart/pyautogui

    work page 2025

  50. [50]

    The Effects of Electrical Stimulation Parameters in Managing Spasticity After Spinal Cord Injury: A Systematic Review,

    A. H. Bekhet, V. Bochkezanian, I. M. Saab, and A. S. Gorgey, “The Effects of Electrical Stimulation Parameters in Managing Spasticity After Spinal Cord Injury: A Systematic Review,” Am J Phys Med Rehabil, vol. 98, no. 6, pp. 484–499, Jun. 2019, doi: 10.1097/PHM.0000000000001064

    work page doi:10.1097/phm.0000000000001064 2019

  51. [51]

    Effect of frequency and pulse duration on human muscle fatigue during repetitive electrical stimulation,

    T. Kesar and S. Binder‐Macleod, “Effect of frequency and pulse duration on human muscle fatigue during repetitive electrical stimulation,” Experimental Physiology , vol. 91, no. 6, pp. 967 –976, Nov. 2006, doi: 10.1113/expphysiol.2006.033886

    work page doi:10.1113/expphysiol.2006.033886 2006

  52. [52]

    Functional electrical stimulation therapy for restoration of motor function after spinal cord injury and stroke: a review,

    C. Marquez-Chin and M. R. Popovic, “Functional electrical stimulation therapy for restoration of motor function after spinal cord injury and stroke: a review,” BioMed Eng OnLine, vol. 19, no. 1, p. 34, Dec. 2020, doi: 10.1186/s12938-020-00773-4

    work page doi:10.1186/s12938-020-00773-4 2020

  53. [53]

    Evaluating the health and fitness benefits of a 6 -month FES-cycling program on a recumbent trike for individuals with motor complete SCI: a pilot study,

    N. Sanna et al., “Evaluating the health and fitness benefits of a 6 -month FES-cycling program on a recumbent trike for individuals with motor complete SCI: a pilot study,” J NeuroEngineering Rehabil , vol. 22, no. 1, p. 55, Mar. 2025, doi: 10.1186/s12984-025-01585-0

    work page doi:10.1186/s12984-025-01585-0 2025

  54. [54]

    Body composition changes after 12 months of FES cycling: case report of a 60 -year-old female with paraplegia,

    D. R. Dolbow, A. S. Gorgey, D. R. Gater, and J. R. Moore, “Body composition changes after 12 months of FES cycling: case report of a 60 -year-old female with paraplegia,” Spinal Cord , vol. 52, no. S1, pp. S3 –S4, Jun. 2014, doi: 10.1038/sc.2014.40

    work page doi:10.1038/sc.2014.40 2014