common-synaptic-input-estimation-hdemg

name	common-synaptic-input-estimation-hdemg
description	Practical methodology for estimating common synaptic input to spinal motor neurons from high-density surface EMG motor unit spike trains using openhdemg. Three complementary approaches: time-domain, frequency-domain, and network-information methods.
category	neuroscience
tags	["EMG","motor-neurons","synaptic-input","HDsEMG","openhdemg","coherence","graph-theory","motor-control","BCI"]
date_created	2026-06-29
arxiv_id	2606.23066v1
authors	["Helio V. Cabral","Giacomo Valli","Roberto Zanotti","Ioannis Delis","Francesco Negro"]
activation	common synaptic input estimation, motor unit decomposition, HDsEMG analysis, openhdemg, coherence analysis, motor neuron control

Common Synaptic Input Estimation from Motor Unit Spike Trains

Overview

Common synaptic input (CSI) represents the dominant component of neural drive transmitted from motor neurons to muscle. This methodology provides a practical, physiologically grounded guide for estimating CSI from populations of motor unit (MU) spike trains using openhdemg, an open-source Python framework.

Three Complementary Approaches

1. Time-Domain: Smoothed Discharge Rates

• Principle: Apply smoothing to instantaneous discharge rates, then compute correlation
• Physiological Interpretation: Captures shared low-frequency fluctuations in motor neuron drive
• Key Parameters: Smoothing window width (critical choice), epoch duration
• Implementation: openhdemg.csi.time_domain()

2. Frequency-Domain: Cumulative Spike Train Coherence

• Principle: Compute coherence between cumulative spike trains (CST)
• Physiological Interpretation: Identifies frequency-specific common oscillatory inputs
• Key Parameters: Segment length, overlap, frequency bands of interest
• Implementation: openhdemg.csi.coherence()
• Bands: Typically 0-5 Hz (common drive), 5-20 Hz (beta), 20-40 Hz (gamma)

3. Network-Information: Pairwise Dependencies + Graph Theory

• Principle: Nonlinear pairwise dependencies modeled as graph edges
• Physiological Interpretation: Reveals network topology of shared inputs
• Key Parameters: Dependency measure (mutual information, transfer entropy), graph threshold
• Implementation: openhdemg.csi.network()

Methodology Workflow

Step 1: HDsEMG Decomposition

import openhdemg

# Load HDsEMG data
data = openhdemg.load("emg_recording.h5")

# Decompose into motor unit spike trains
mus = openhdemg.decompose(data, method="fastica")
# Returns: list of MotorUnit objects with spike_trains attribute

Step 2: Motor Unit Cleaning

mus_clean = openhdemg.filter_mu(mus, 
    min_discharges=,          
    maxisi_ratio=,           
    min_sil=)