Wearable EMG MUAP Decoding Using Neuromorphic Spike Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Accurate measurement of electromyography (EMG) signals is challenging due to artifacts introduced by electrode impedance changes, radio frequency interference, and interference from neuromuscular electrical stimulation (NMES) pulses, which can distort the EMG signal.
Innovation Solution
A wearable EMG measurement device with integrated electrodes and electronics that perform motor unit action potential (MUAP) decomposition using analog matrix processors, delta sigma converters, and neuromorphic chips for blind source separation and spiking neural networks to process EMG data, enhancing signal isolation and decoding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If EMG measurements are interleaved with NMES pulses, then EMG-guided NMES applications can be implemented, but the NMES pulses introduce interference that distorts the EMG signal measurement
Solution Approach 1:
The patent segments the EMG signal processing into multiple independent channels, each processed through separate blind source separation algorithms. This segmentation allows the system to isolate and process different signal sources (EMG vs. NMES interference) independently, resolving the contradiction by maintaining measurement accuracy while enabling guided NMES applications
Solution Approach 2:
The patent introduces blind source separation as an intermediary processing step between signal acquisition and analysis. This intermediary technique separates mixed signals into their original sources, allowing the system to extract clean EMG signals even when NMES pulses are present, thus maintaining reliability while enabling application versatility
2Productivity
If conventional digital signal processing is used for MUAP decomposition, then processing can be performed, but power consumption and computational latency increase
Solution Approach 1:
The patent replaces conventional digital signal processing (mechanical/computational system) with neuromorphic computing that mimics biological neural processing. This substitution enables real-time MUAP decomposition with significantly reduced power consumption, as the neuromorphic system processes signals in parallel event-driven manner rather than through power-intensive sequential digital computation
3Area of stationary object
If multiple electrodes are used to improve EMG signal coverage, then measurement coverage increases, but susceptibility to artifacts from electrode impedance changes and RFI increases
Solution Approach 1:
The patent extracts and removes artifact components from the multi-channel EMG signal using blind source separation. By separating the mixed signals into their original sources, the system can identify and eliminate artifacts from electrode impedance changes and radio frequency interference while retaining the useful EMG signal components, thus maintaining wide coverage while reducing artifact susceptibility
Data Source
AI summary
A garment is worn on an anatomical region, with electrodes arranged on the garment to contact skin of the anatomical region. An EMG amplifier measures analog EMG data, and analog circuitry decomposes the analog EMG data into MUAPs. The analog circuitry may include an analog matrix processor performing analog matrix multiplication to transform the analog EMG data into source signals, an analog squarer circuit computing power signals by squaring the source signals, and delta sigma analog-to-digital converters converting the power signals to analog spike signals. The analog circuitry may include a neuromorphic chip to transform the analog EMG data into analog spike signals using blind source separation. A neuromorphic chip may process the analog spike signals to determine volitional intent using spiking neural networks (SNN), and/or perform a neuromuscular debilitation assessment based on the analog spike signals or encoded spike train