Abstract
The study of human neuromechanical control at the motor unit (MU) level has predominantly focussed on electrical activity and force generation, whilst the link between these, the muscle deformation, has not been widely studied. We combine high density surface electromyography (HDsEMG) and ultrafast ultrasound (US) recordings of a mildly contracted muscle (tibialis anterior) to measure the deformation of the muscular tissue caused by individual MU twitches. With a novel analysis on the US images we identified the velocity maps associated with single muscle unit movements. The excitation-contraction (E-C) coupling was 3.8 ± 3.0 ms, providing the first measurement of this value for single MUs in non-stimulated contractions. We also found evidence of single muscle unit twisting and split territories. This approach opens new avenues for study of neuromechanics in healthy and pathological conditions.
Introduction: The "Black Box" of Muscle Contraction
In neuromuscular research, scientists typically observe the electrical "spark" (neural drive via EMG) and the final mechanical "output" (global force). However, the intermediate physical deformation of individual muscle fibers during natural movement has largely remained a "black box." A critical biomarker in this process is excitation-contraction (E-C) coupling—the millisecond-scale delay where an electrical signal triggers physical tension. Historically, measuring E-C coupling in vivo required artificial electrical stimulation, which does not reflect natural neural control. This study introduces a non-invasive method to observe these individual mechanical events during voluntary contractions.
Methods: A Multimodal Imaging Pipeline
The researchers developed a pipeline combining two high-resolution technologies to isolate the mechanical "fingerprint" of single motor units (MUs):
- High-Density Surface Electromyography (HDsEMG): 128 electrodes (two 64-electrode grids) were used to detect the electrical firing patterns of individual MUs.
- Ultrafast Ultrasound (US): Capturing 1000 frames per second, this allowed for the tracking of minute tissue velocity changes in the tibialis anterior muscle.
The core of the analysis is Spike-Triggered Averaging (STA). By using the precise firing times from the EMG as a trigger, the team averaged thousands of ultrasound frames to filter out background noise and isolate the velocity map of a specific motor unit. A pixel-wise variability weighting was applied to ensure the resulting profiles were representative of the target unit's consistent movement. The study analyzed 51 distinct motor units from 12 healthy participants during low-force (10% maximum) voluntary contractions.
Key Insights: Twisting and Timing
The integration of electrical and mechanical data yielded several groundbreaking observations:
- Natural E-C Coupling: The average E-C coupling time was measured at 3.8 ± 3.0 ms. This is the first time this metric has been captured during non-stimulated, voluntary contractions.
- Mechanical Twisting: Approximately 17% of the motor units exhibited "twisting" behavior, where different parts of the same unit moved in opposite directions relative to the ultrasound probe. This suggests that muscle unit architecture is significantly more complex than simple linear models.
- Motion Domains: The researchers distinguished between the anatomical "MU territory" and the "MU motion domain" (the region of tissue actually displaced). These domains were found to be highly repeatable across trials but could be "split" into multiple distinct regions across the muscle.
Quantitative Results
The study provided high-precision temporal metrics for single muscle twitches:
- Mechanical Twitch Duration: 17.9 ± 5.3 ms (active tissue acceleration).
- Active Contraction Time: 12.1 ± 4.0 ms.
- Total Contraction Time: 56.6 ± 8.4 ms.
- Spatial Consistency: Motor units showed 47% spatial overlap across trials, confirming the method's ability to reliably isolate individual units compared to the 7% overlap seen between different units.
Limitations and Future Directions
While revolutionary, the method currently has limitations. It is restricted to low-force levels (up to 10% MVC) because higher forces cause global muscle stiffness that masks individual unit vibrations. Additionally, the use of surface EMG biases the data toward units near the skin. Future work aims to move from 1D axial tracking to 2D or 3D vector flow imaging to fully capture the complex twisting and expansion of muscle tissue. This research paves the way for new diagnostic tools capable of detecting mechanical markers of neuromuscular diseases long before global functional loss occurs.
Summary updated