EMG-projected MEG high-resolution source imaging of human motor execution: Brain-muscle coupling above movement frequencies

Mingxiong Huang, Deborah L. Harrington, Annemarie Angeles‐Quinto, Zhengwei Ji, Ashley Robb‐Swan, Charles Huang, Qian Shen, Hayden Hansen, J. D. Baumgartner, Jaqueline Hernandez-Lucas, Sharon Nichols, Joanna Jacobus, Tao Song, Imanuel Lerman, Maksim Bazhenov, Giri Krishnan, Dewleen G. Baker, Ramesh Rao, Roland R. Lee

TL;DR

Researchers have developed a novel "EMG-projected MEG" technique that can map the human primary motor cortex (M1) in as little as 60 seconds. By mathematically projecting brain activity onto muscle signals, this method bypasses the need for hundreds of repetitive trials, making high-resolution functional mapping feasible for patients with brain tumors or strokes who struggle with traditional motor tasks.

What Problem Does This Paper Address?

Before a neurosurgeon removes a brain tumor, they need a "road map" to avoid damaging the primary motor cortex (M1), the region responsible for voluntary movement. Currently, the gold standards are fMRI or traditional Magnetoencephalography (MEG).

However, these methods have a major flaw: they rely on "trial averaging." To get a clear signal, a patient must perform the same movement (like a finger tap) hundreds of times in a perfectly timed, rhythmic fashion. For patients with brain lesions, weakness, or tremors, this is often impossible. Their movements are frequently slow, weak, or irregular, leading to "noisy" data that fails to localize the motor strip.

Furthermore, there is a scientific knowledge gap regarding Corticomuscular Coupling (CMC)—how the brain and muscles communicate at frequencies higher than the actual physical movement. Huang et al. sought to bridge this gap with a faster, more robust imaging solution.

Methods

The authors introduced a technique called EMG-projected MEG source imaging.

  1. Data Collection: Participants performed self-paced index finger flexions (~1 Hz) for about one minute while MEG (brain) and EMG (muscle) signals were recorded simultaneously.
  2. The "Projection" Innovation: Instead of averaging trials based on time, the researchers used an enhanced version of the Fast-VESTAL algorithm. They mathematically projected the MEG sensor data into a parameter space defined by the EMG activity.
  3. Temporal Filtering: This projection acts as a powerful filter. It amplifies brain signals that are specifically "talking" to the muscles while suppressing background brain noise and artifacts.
  4. Frequency Decomposition: They analyzed five distinct frequency bands: Delta (1-4 Hz), Theta (4-7 Hz), Alpha (8-12 Hz), Beta (15-30 Hz), and Gamma (30-90 Hz) to see which provided the most accurate map.

Datasets

The study evaluated the method using:

  • Healthy Controls: 13 right-handed participants (yielding 26 datasets for left and right hand movements).
  • Clinical Patients: Two presurgical patients with significant sensorimotor dysfunction.
  • Patient 1 had a metastatic melanoma near the motor strip.
  • Patient 2 had a perinatal stroke and "highly irregular" movement patterns that had previously caused a 3-Tesla fMRI mapping attempt to fail.

Key Insights and Ideas

  • The "Theta" Sweet Spot: The most striking finding was that the Theta band (4-7 Hz) provided 100% accuracy in localizing M1. Because this frequency is higher than the physical movement (~1 Hz), it suggests that theta-band coupling is driven by the brain's internal motor control rather than just the physical rhythm of the finger.
  • Independence from Rhythm: Unlike traditional methods, this approach doesn't care if the movement is rhythmic. In Patient 2, whose EMG was chaotic and lacked a fundamental frequency, the theta-band projection still accurately pinpointed the motor cortex.
  • Speed vs. Accuracy: The researchers found they could generate accurate maps using only the first 30 seconds of data (96.2% accuracy) and even saw success with just 5 seconds of data (73.1% accuracy).
  • Why Alpha and Gamma Failed: The alpha band showed poor coupling (34.6%), likely due to "mu rhythm" suppression during movement. Gamma band coupling was non-existent (0.0%), possibly due to the lower signal-to-noise ratio of non-invasive MEG at high frequencies.

Key Results

  • Localization Accuracy: In healthy subjects, the method achieved 100% accuracy in the Delta and Theta bands, and 76.9% in the Beta band.
  • Clinical Success: In both surgical patients, the technique successfully localized M1. For Patient 1, this map allowed surgeons to perform a total resection of a tumor while successfully avoiding the motor strip, resulting in no neurological deficits post-surgery.
  • Efficiency: The task requires ~1 minute of recording, compared to the 10-20 minutes required by conventional MEG movement paradigms.

Limitations and Future Work

The authors acknowledge that while the results are promising, the clinical sample size is small (n=2). More case studies are required to fully validate the approach across diverse pathologies. Additionally, the failure of the gamma band suggests that non-invasive MEG may still struggle with very high-frequency brain-muscle communication compared to invasive methods like ECoG (Electrocorticography).

Future research could explore whether this "projection" approach can be applied to other functional areas, such as language or sensory mapping, further reducing the burden on patients during presurgical workups.

Summary updated

Abstract

Magnetoencephalography (MEG) is a non-invasive functional imaging technique for pre-surgical mapping. However, movement-related MEG functional mapping of primary motor cortex (M1) has been challenging in presurgical patients with brain lesions and sensorimotor dysfunction due to the large numbers of trials needed to obtain adequate signal to noise. Moreover, it is not fully understood how effective the brain communication is with the muscles at frequencies above the movement frequency and its harmonics. We developed a novel Electromyography (EMG)-projected MEG source imaging technique for localizing early-stage (-100 to 0 ms) M1 activity during ~l min recordings of left and right self-paced finger movements (~1 Hz). High-resolution MEG source images were obtained by projecting M1 activity towards the skin EMG signal without trial averaging. We studied delta (1-4 Hz), theta (4-7 Hz), alpha (8-12 Hz), beta (15-30 Hz), gamma (30-90 Hz), and upper-gamma (60-90 Hz) bands in 13 healthy participants (26 datasets) and three presurgical patients with sensorimotor dysfunction. In healthy participants, EMG-projected MEG accurately localized M1 with high accuracy in delta (100.0%), theta (100.0%), and beta (76.9%) bands, but not alpha (34.6%) or gamma/upper-gamma (0.0%) bands. Except for delta, all other frequency bands were above the movement frequency and its harmonics. In three presurgical patients, M1 activity in the affected hemisphere was also accurately localized, despite highly irregular EMG movement patterns in one patient. Altogether, our EMG-projected MEG imaging approach is highly accurate and feasible for M1 mapping in presurgical patients. The results also provide insight into movement-related brain-muscle coupling above the movement frequency and its harmonics.
BibTeX
@inproceedings{huang2023emg,
  author = {Mingxiong Huang and Deborah L. Harrington and Annemarie Angeles‐Quinto and Zhengwei Ji and Ashley Robb‐Swan and Charles Huang and Qian Shen and Hayden Hansen and J. D. Baumgartner and Jaqueline Hernandez-Lucas and Sharon Nichols and Joanna Jacobus and Tao Song and Imanuel Lerman and Maksim Bazhenov and Giri Krishnan and Dewleen G. Baker and Ramesh Rao and Roland R. Lee},
  title = {{EMG-projected MEG high-resolution source imaging of human motor execution: Brain-muscle coupling above movement frequencies}},
  booktitle = {Imaging},
  year = {2023},
  doi = {10.1162/imag\_a\_00056},
}

Loading related content...

Figure