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

TL;DR

MyoKin3X is a customizable, validated full-hand 3D force measurement system combining sensors and software for high-fidelity hand motor control studies.

Contribution

It introduces a versatile framework with integrated sensors and software enabling reliable, standardized, and multi-digit force measurement and analysis.

Findings

High stability of calibration factors with 0.04% variation.

Effective inter-axis decoupling with 92.71% crosstalk reduction.

Predictive accuracy with R2 > 0.99 across sensors.

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.