pith. sign in

arxiv: 2604.27949 · v1 · submitted 2026-04-30 · 💻 cs.HC

MyoKin3X: A Myoelectric Framework for Full-Hand 3D Force Recording

Charlotte Rohleder , Raul S\^impetru , Annika W\"unsch , Alessandro Del Vecchio This is my paper

Pith reviewed 2026-05-07 06:32 UTC · model grok-4.3

classification 💻 cs.HC
keywords myoelectric control3D force measurementhand coordinationfinger forcessensor calibrationEMG synchronizationmotor controlhuman-machine interface
0
0 comments X

The pith

MyoKin3X enables simultaneous 3D force recording from all five digits with validated stability and low crosstalk.

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

The paper introduces MyoKin3X as a customizable hardware and software system for measuring three-dimensional forces from up to five fingers at the same time while also capturing muscle activity signals. It tackles the difficulty of getting accurate multi-directional force data across the whole hand by pairing an adaptable mechanical frame with five built-in 3D sensors and a calibration routine that corrects for interference between axes and fingers. A sympathetic reader would care because reliable full-hand force data could support clearer experiments on how people coordinate finger actions, how muscles work in groups, and how to build better links between the body and machines. The reported tests show calibration factors hold steady with almost no drift and that force predictions stay accurate after removing most cross-talk between measurement directions.

Core claim

MyoKin3X combines an anatomically versatile structure with five integrated 3D force sensors and a standalone software package for synchronized electromyography and force acquisition. It supplies in-place cross-calibration of all sensors, records single- and multi-digit maximal contractions, and applies automated coordinate transformations to produce digit-specific data for comparison across people and tasks. Validation shows calibration factors with a mean coefficient of variation of 0.04 percent, maximum force error of plus or minus 0.06 newtons at 50 newtons, mean crosstalk reduction of 92.71 percent, residual crosstalk below 0.02 percent for most axis pairs, and predictive accuracy with R

What carries the argument

The anatomically versatile structure holding five 3D force sensors together with the in-place cross-calibration routine and automated transformation to digit-specific coordinate systems.

Where Pith is reading between the lines

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

  • The system could support clinical assessments of hand function in people recovering from injury or stroke by providing standardized multi-finger force profiles.
  • High-accuracy force data from this framework might be fed into machine-learning models to create more responsive myoelectric prosthetics that adjust to individual finger loads.
  • Extending recordings to natural, unconstrained movements outside the lab would test whether the low crosstalk persists when grip patterns vary freely.
  • Pairing the sensors with wireless transmission could allow longer sessions that capture how force coordination changes with fatigue in everyday activities.

Load-bearing premise

The adaptable mechanical structure and cross-calibration procedure will deliver the reported accuracy, stability, and low crosstalk consistently for hands of different sizes and during tasks outside the specific validation conditions.

What would settle it

Repeating the full calibration validation on a new group of twenty subjects whose hand lengths span the 5th to 95th percentiles and finding a mean coefficient of variation above 0.5 percent or residual crosstalk above 1 percent in any axis pair would show the performance does not hold.

Figures

Figures reproduced from arXiv: 2604.27949 by Alessandro Del Vecchio, Annika W\"unsch, Charlotte Rohleder, Raul S\^impetru.

Figure 1
Figure 1. Figure 1: Framework Overview. A) 1D ramp feedback showing target and exerted task-related force magnitude (digit component breakdown shown in main setup sub-figure for clarity). B) 1D fatigue feedback showing exerted magnitude as a bar, target as a black dotted line, and acceptable deviation range as two grey lines. C) Drawing of the MyoKin3X setup with the 3D force framework and the visual interface displaying elec… view at source ↗
Figure 2
Figure 2. Figure 2: Force Framework Setup Overview and Components. A) Top computer-aided design (CAD) view with five mounted sensors and two arm-support profiles. B) Front-left CAD view of the force framework with force sensor and intramuscular EMG amplifier holder. C) CAD view of one sensor assembly with digit box and telescopic inlets of different size. D) Close-up CAD view with two digit boxes mounted on different sides of… view at source ↗
Figure 3
Figure 3. Figure 3: Sensor Decoupling Effectiveness. Crosstalk analysis for one sensor showing inter-axis coupling before and after decoupling. A) Before decoupling, inter-axis coupling ranges from 0.381% to 3.602% on off￾diagonal elements. B) After decoupling, residual crosstalk is reduced to 0%–0.011%, while the worst-case residual across all sensors is 0.175%. dimensions and are fixed by a rear screw. Two front hori￾zontal… view at source ↗
read the original abstract

Simultaneous multi-directional force measurement across all five digits is essential for studying hand coordination, compensatory forces, and myoelectric control, yet existing systems trade off digit coverage, force dimensionality, and anatomical adaptability. Reliable full-hand acquisition remains challenging because multi-axis calibration, hand-size adjustment, and consistent digit-specific force reconstruction are technically demanding. We present MyoKin3X, a customizable full-hand framework for simultaneous 3D force measurement of up to five digits providing robust and validated force reconstruction. It combines an anatomically versatile structure with five integrated 3D force sensors and a standalone software for synchronized electromyography and force acquisition. MyoKin3X provides in-place cross-calibration of all five sensors, single- and multi-digit maximal voluntary contraction recording, and automated coordinate transformation to digit-specific coordinate systems for standardized analysis across subjects and tasks. Calibration validation demonstrates high stability of the axis-specific calibration factors, with a mean coefficient of variation of 0.04% and maximum force error of +- 0.06N at 50N. It also shows effective inter-axis decoupling (mean crosstalk reduction: 92.71%; residual crosstalk below 0.02% for most axis pairs) and high predictive accuracy (R2 > 0.99 across sensors). The software includes four feedback modes: 1D ramps, fatigue protocols, 2D arbitrary target ramps, and 2D exploratory tasks. MyoKin3X therefore enables standardized full-hand force acquisition with validated measurement reliability, flexible protocol control, and real-time visualization for high-fidelity studies of hand motor control, muscle synergies, and human-machine interfacing.

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

1 major / 1 minor

Summary. The manuscript presents MyoKin3X, a customizable full-hand framework for simultaneous 3D force measurement across up to five digits. It integrates five 3D force sensors on an anatomically versatile structure with standalone software for synchronized EMG and force acquisition, in-place cross-calibration, single- and multi-digit MVC recording, automated coordinate transformation to digit-specific systems, and four feedback modes (1D ramps, fatigue protocols, 2D target ramps, 2D exploratory tasks). The central claims rest on calibration validation results: mean coefficient of variation of 0.04% for axis-specific factors, maximum force error of ±0.06 N at 50 N, mean crosstalk reduction of 92.71% with residual crosstalk below 0.02% for most axis pairs, and R² > 0.99 predictive accuracy across sensors, supporting robust and standardized force reconstruction for hand motor control studies.

Significance. If the reported stability, low crosstalk, and accuracy generalize reliably across subjects and hand sizes, MyoKin3X would address a genuine gap in existing systems by enabling anatomically adaptable, multi-digit 3D force recording with integrated myoelectric synchronization and real-time feedback. The in-place cross-calibration approach and software features represent practical engineering contributions that could support reproducible protocols in HCI, rehabilitation, and motor neuroscience. However, the current evidence base for these claims is limited by the absence of multi-subject validation details, which weakens the extrapolation to 'standardized full-hand force acquisition' across users.

major comments (1)
  1. [Abstract / Calibration validation] Abstract, calibration validation paragraph: The performance metrics (mean CV 0.04%, max error ±0.06 N at 50 N, 92.71% crosstalk reduction, R² > 0.99) are presented as demonstrating 'robust and validated force reconstruction' and 'anatomically versatile structure,' yet no subject count, hand-size range, or test of sensor repositioning/re-calibration for different anatomies is reported. This directly undermines the load-bearing claim that the framework delivers consistent decoupling and stability across diverse users, as the validation appears confined to a single fixed setup without demonstrated generalizability.
minor comments (1)
  1. [Abstract] The abstract would be strengthened by explicitly stating the number of participants and hand-size demographics in the validation section to allow readers to assess the scope of the reported metrics.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their thorough review and valuable feedback on our manuscript. We agree that the abstract's presentation of the calibration metrics could be clarified to more precisely reflect the scope of the single-setup validation performed. We have revised the manuscript accordingly and provide the following point-by-point response.

read point-by-point responses

  1. Referee: [Abstract / Calibration validation] Abstract, calibration validation paragraph: The performance metrics (mean CV 0.04%, max error ±0.06 N at 50 N, 92.71% crosstalk reduction, R² > 0.99) are presented as demonstrating 'robust and validated force reconstruction' and 'anatomically versatile structure,' yet no subject count, hand-size range, or test of sensor repositioning/re-calibration for different anatomies is reported. This directly undermines the load-bearing claim that the framework delivers consistent decoupling and stability across diverse users, as the validation appears confined to a single fixed setup without demonstrated generalizability.

    Authors: We appreciate the referee highlighting this important point. The calibration validation was performed on a single fixed setup to characterize the stability of the axis-specific calibration factors (mean CV of 0.04%), force reconstruction accuracy (maximum error ±0.06 N at 50 N), inter-axis decoupling via in-place cross-calibration (mean crosstalk reduction 92.71%, residual below 0.02% for most pairs), and predictive accuracy (R² > 0.99). These results specifically validate the technical reliability of the sensor array, cross-calibration procedure, and automated coordinate transformation within the tested configuration. The anatomical versatility is a core design feature of the adjustable structure, and the software supports per-session recalibration for different users or hand sizes. We acknowledge that the original abstract phrasing could be read as implying broader generalizability across subjects and anatomies than the single-setup data supports. To address this, we have revised the abstract to state that the metrics demonstrate robust calibration and decoupling performance in the reported system, rather than claiming standardized acquisition across diverse users. We have also added a paragraph in the discussion section noting that while the framework is designed for adaptability, empirical multi-subject validation across hand sizes and repositioning tests remains an important direction for future work. These changes ensure the claims are precisely supported by the presented evidence. revision: yes

Circularity Check

0 steps flagged

No circularity: validation metrics obtained from external reference forces

full rationale

The paper reports empirical calibration outcomes (mean CV of 0.04%, max error ±0.06 N at 50 N, 92.71% crosstalk reduction, R² > 0.99) obtained via in-place cross-calibration against independent reference loads. No equations, self-citations, or fitted parameters are presented that would make these quantities equivalent to their inputs by construction. The central claims rest on measured performance of the hardware-software system rather than any self-definitional loop or renamed ansatz. Generalizability across hand sizes is an untested empirical question but does not constitute circular reasoning in the derivation chain.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

This is an engineering development paper describing a physical device and software system. The central claims rest on empirical hardware design, experimental calibration procedures, and reported validation metrics rather than mathematical axioms, free parameters, or invented theoretical entities.

pith-pipeline@v0.9.0 · 5614 in / 1285 out tokens · 107507 ms · 2026-05-07T06:32:56.036135+00:00 · methodology

discussion (0)

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

Reference graph

Works this paper leans on

21 extracted references · 11 canonical work pages

  1. [1]

    Instrumented platform for assessment of isometric hand muscles contractions,

    N. Male ˇsevi´c, G. Andersson, A. Bj ¨orkman, M. Controzzi, C. Cipriani, and C. Antfolk, “Instrumented platform for assessment of isometric hand muscles contractions,”Measurement Science and Technology, vol. 30, no. 6, 4 2019

    2019

  2. [2]

    F. E. Zajac, “How musculotendon architecture and joint geometry affect the capacity of muscles to move and exert force on objects: A review with application to arm and forearm tendon transfer design,” The Journal of Hand Surgery, vol. 17, no. 5, pp. 799–804, 9

  3. [3]

    Available: https://linkinghub.elsevier.com/retrieve/pii/ 036350239290445U

    [Online]. Available: https://linkinghub.elsevier.com/retrieve/pii/ 036350239290445U

  4. [4]

    Precision Grip in Humans,

    M.-C. Hepp-Reymond, E. J. Huesler, and M. A. Maier, “Precision Grip in Humans,” inHand and Brain. Elsevier, 1996, pp. 37–68. [Online]. Available: https://linkinghub.elsevier.com/retrieve/ pii/B9780127594408500062

    1996

  5. [5]

    Tutorial: Analysis of motor unit discharge characteristics from high-density surface EMG signals,

    A. Del Vecchio, A. Holobar, D. Falla, F. Felici, R. Enoka, and D. Farina, “Tutorial: Analysis of motor unit discharge characteristics from high-density surface EMG signals,”Journal of Electromyography and Kinesiology, vol. 53, p. 102426, 8 2020. [On- line]. Available: https://doi.org/10.1016/j.jelekin.2020.102426https: //linkinghub.elsevier.com/retrieve/p...

    work page doi:10.1016/j.jelekin.2020.102426https: 2020

  6. [6]

    Incomplete Functional Subdivision of the Human Multitendoned Finger Muscle Flexor Digitorum Profundus: An Electromyographic Study,

    K. T. Reilly and M. H. Schieber, “Incomplete Functional Subdivision of the Human Multitendoned Finger Muscle Flexor Digitorum Profundus: An Electromyographic Study,”Journal of Neurophysiology, vol. 90, no. 4, pp. 2560–2570, 10 2003. [Online]. Available: https://www.physiology.org/doi/10.1152/jn.00287.2003

    work page doi:10.1152/jn.00287.2003 2003

  7. [7]

    Intramuscular High-Density Micro- Electrode Arrays Enable High-Precision Decoding and Mapping of Spinal Motor Neurons to Reveal Hand Control,

    A. Grison, J. I. Pereda, S. Muceli, A. Kundu, F. Baracat, G. Indiveri, E. Donati, and D. Farina, “Intramuscular High-Density Micro- Electrode Arrays Enable High-Precision Decoding and Mapping of Spinal Motor Neurons to Reveal Hand Control,” 10 2024. [Online]. Available: http://arxiv.org/abs/2410.11016

    work page arXiv 2024

  8. [8]

    A Device for Testing the Intrinsic Muscles of the Hand,

    T. C. Pataky, A. V . Savescu, M. L. Latash, and V . M. Zatsiorsky, “A Device for Testing the Intrinsic Muscles of the Hand,”Journal of Hand Therapy, vol. 20, no. 4, pp. 345–350, 10 2007. [Online]. Available: https://linkinghub.elsevier.com/retrieve/pii/S0894113007000610

    2007

  9. [9]

    Development and application of a multi-axis dynamometer for measuring grip force,

    C. Irwin, J. Towles, and R. Radwin, “Development and application of a multi-axis dynamometer for measuring grip force,”Ergonomics, vol. 56, no. 12, pp. 1841–1849, 12 2013. [Online]. Available: http://www.tandfonline.com/doi/abs/10.1080/00140139.2013.847212

    work page doi:10.1080/00140139.2013.847212 2013

  10. [10]

    A wearable low-cost device based upon Force-Sensing Resistors to detect single-finger forces,

    C. Castellini and V . Ravindra, “A wearable low-cost device based upon Force-Sensing Resistors to detect single-finger forces,” in5th IEEE RAS/EMBS International Conference on Biomedical Robotics and Biomechatronics. IEEE, 8 2014, pp. 199–203. [Online]. Available: https://ieeexplore.ieee.org/document/6913776

    work page arXiv 2014

  11. [11]

    Design of a Custom Force Fixture Fitting Miniature Load Cells for Individual Finger Force Measurement,

    L. Ross, N. Kutchey, N. Rayess, and M. O. Conrad, “Design of a Custom Force Fixture Fitting Miniature Load Cells for Individual Finger Force Measurement,”Proceedings of the Human Factors and Ergonomics Society Annual Meeting, vol. 67, no. 1, pp. 379–384, 9

  12. [12]

    Available: https://journals.sagepub.com/doi/10.1177/ 21695067231192434

    [Online]. Available: https://journals.sagepub.com/doi/10.1177/ 21695067231192434

  13. [13]

    Finger interaction in a three-dimensional pressing task,

    S. Kapur, J. Friedman, V . M. Zatsiorsky, and M. L. Latash, “Finger interaction in a three-dimensional pressing task,”Experimental Brain Research, vol. 203, no. 1, pp. 101–118, 5 2010. [Online]. Available: http://link.springer.com/10.1007/s00221-010-2213-7

    work page doi:10.1007/s00221-010-2213-7 2010

  14. [14]

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

    R. C. S ˆımpetru, 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 Transactions on Neural Systems and Rehabilitation Engineering, vol. 32, pp. 3741–3750, 2024. [Online]. Available: https://ieeexpl...

    work page arXiv 2024

  15. [15]

    Proportional and Simultaneous Real-Time Control of the Full Human Hand From High- Density Electromyography,

    R. C. S ˆımpetru, M. M ¨arz, and A. Del Vecchio, “Proportional and Simultaneous Real-Time Control of the Full Human Hand From High- Density Electromyography,”IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 31, pp. 3118–3131, 2023. [Online]. Available: https://ieeexplore.ieee.org/document/10182328/

    work page arXiv 2023

  16. [16]

    Biosignal-device-interface,

    D. Braun and R. Simpetru, “Biosignal-device-interface,”

  17. [17]

    Available: https://github.com/NsquaredLab/ Biosignal-Device-Interface

    [Online]. Available: https://github.com/NsquaredLab/ Biosignal-Device-Interface

  18. [18]

    Multiaxis Force/Torque Sensor Technologies: Design Principles and Robotic Force Control Applications: A Review,

    S. Li and J. Xu, “Multiaxis Force/Torque Sensor Technologies: Design Principles and Robotic Force Control Applications: A Review,”IEEE Sensors Journal, vol. 25, no. 3, pp. 4055–4069, Feb. 2025. [Online]. Available: https://ieeexplore.ieee.org/document/10755023

    work page arXiv 2025

  19. [19]

    Decoupled Six-Axis Force-Moment Sensor with a Novel Strain Gauge Arrange- ment and Error Reduction Techniques,

    G. A. Kebede, A. R. Ahmad, S.-C. Lee, and C.-Y . Lin, “Decoupled Six-Axis Force-Moment Sensor with a Novel Strain Gauge Arrange- ment and Error Reduction Techniques,”Sensors, vol. 19, no. 13, p. 3012, Jul. 2019

    2019

  20. [20]

    Design, Development and Calibration of a Lightweight, Compliant Six-Axis Optical Force/Torque Sensor,

    O. Al-Mai, M. Ahmadi, and J. Albert, “Design, Development and Calibration of a Lightweight, Compliant Six-Axis Optical Force/Torque Sensor,”IEEE Sensors Journal, vol. 18, no. 17, pp. 7005–7014, Sep. 2018. [Online]. Available: https://ieeexplore.ieee. org/abstract/document/8410875

    work page arXiv 2018

  21. [21]

    Reliability and validity of the finger flexor dynamometer,

    K. Nestler, U. Rohde, B. Becker, S. Waldeck, D. A. Veit, and D. Leyk, “Reliability and validity of the finger flexor dynamometer,”Hand Therapy, vol. 24, no. 3, pp. 82–90, Sep. 2019. [Online]. Available: https://journals.sagepub.com/doi/10.1177/1758998319859382

    work page doi:10.1177/1758998319859382 2019