Ain Shams University
December 17,2023
Lab 3
Submitted by
• Abdelrahman Mohamed Hagag 20P7509
• Mohamed Ahmed Fathy 20P7427
• Mostafa Ibrahim Ali 20P3368 Introduction to Bio-
• Youssef Walied Barakat 20P7049
Mechatronics
• Yahia Ihab 20P8571
Presented to
• Eng Hamdy Osama
Table of Contents
Introduction ...................................................................................................................................... 2
EMG signal acquisition .................................................................................................................... 3
1. Electrode Technology and Placement: .................................................................................. 3
2. High-Fidelity Sensors and Amplification: ............................................................................. 3
3. Multi-channel Systems and Wireless Transmission: ............................................................. 3
Why This Matters: ........................................................................................................................ 3
Better Signals ..................................................................................................................... 3
More Muscle Info .............................................................................................................. 3
Easier Movement ............................................................................................................... 3
EMG signal processing ..................................................................................................................... 4
1. Noise Reduction and Filtering: .............................................................................................. 4
2. Feature Extraction and Analysis: ........................................................................................... 4
3. Pattern Recognition and Machine Learning: ......................................................................... 4
4. Real-time Processing for Prosthetic Control: ........................................................................ 4
5. Optimization and Integration: ................................................................................................ 4
Improving EMG signal acquisition .................................................................................................. 6
1. Advanced Electrode Development: ....................................................................................... 6
2. Optimal Electrode Placement: ............................................................................................... 6
3. Enhanced Sensor Sensitivity: ................................................................................................ 6
4. Advanced Signal Processing Techniques: ............................................................................. 6
5. User Training Initiatives: ....................................................................................................... 6
References ........................................................................................................................................ 7
Figure 1 EMG test _____________________________________________________________ 2
Figure 2 EMG data acquisition ____________________________________________________ 5
Figure 3 EMG signal processing __________________________________________________ 5
[1]
Introduction
EMG signal acquisition taps into muscle electrical activity using electrodes placed either on
the skin's surface or directly within the muscle. There are two main methods:
Surface EMG (sEMG) :
This method involves placing electrodes on the skin, right above the muscle being studied.
It's non-invasive, which means it doesn't require any insertion into the body. However, it's
sensitive to noise and external interferences due to its external placement.
Intramuscular EMG (iEMG) :
In contrast, iEMG requires the insertion of needle-like electrodes directly into the muscle
tissue. While this method provides more accurate and detailed signals due to its proximity
to the muscle fibers, it involves an invasive procedure as it requires penetrating the skin and
muscle tissue. Human-Machine Interaction
Figure 1 EMG test
[2]
EMG signal acquisition
1. Electrode Technology and Placement:
Scientists are working on making better electrodes for EMG. These new ones will give clearer
signals and cause less interference. They're also figuring out the best spots to put these
electrodes on muscles so they can pick up the right signals and avoid mistakes.
2. High-Fidelity Sensors and Amplification:
There's a search for sensors that can detect tiny muscle movements better. They want these
sensors to capture the smallest details in muscle activity. Also, they're using strong
amplifiers to make the EMG signals louder and clearer for easier reading.
3. Multi-channel Systems and Wireless Transmission:
They're making EMG systems that can collect signals from many muscles at once. This helps
to understand how muscles work together. Also, they're working on ways to send EMG
signals wirelessly. This makes it easier for people to move around while their signals are
being recorded.
Why This Matters:
• Better Signals: Improvements in electrodes and sensors mean clearer and more
accurate signals.
• More Muscle Info: Systems that record from many muscles give a better idea of how
our muscles work together.
• Easier Movement: Wireless systems help people move around more freely during
tests, making things more comfortable for them.
[3]
EMG signal processing
To analyze the random signals coming out from the EMG sensor, we first need to pass it through
Some filtration and computational methods :
1. Noise Reduction and Filtering:
We use digital signal processing techniques to filter out unwanted noise and artifacts from EMG
signals. Filtering algorithms help clean the raw signals, ensuring that only relevant muscle activity
information is retained.
2. Feature Extraction and Analysis:
Advanced algorithms are used to extract key features from the processed EMG signals. We identify
and analyze specific signal characteristics such as amplitude, frequency, and time domain
parameters to understand muscle behavior and movement patterns.
3. Pattern Recognition and Machine Learning:
We develop advanced pattern recognition algorithms and machine learning models. These
algorithms interpret the extracted features, recognizing specific muscle activation patterns
associated with particular movements or actions.
4. Real-time Processing for Prosthetic Control:
Real-time processing algorithms are important for immediate interpretation of muscle signals. we
design systems that rapidly process EMG data, allowing prosthetic devices to swiftly respond to user
commands and mimic natural movements.
5. Optimization and Integration:
We constantly optimize signal processing algorithms, aiming for higher accuracy, reduced
computational complexity, and improved prosthetic control. Integration of various sensors, adaptive
algorithms, and feedback mechanisms is key to enhancing the overall functionality of prosthetic
devices.
[4]
Figure 2 EMG data acquisition
Figure 3 EMG signal processing
[5]
Improving EMG signal acquisition
There are already some recommendations discussing on how to improve EMG signal acquisition
1. Advanced Electrode Development:
Researchers are actively engaged in the advancement of electrode materials and designs to optimize
their interaction with muscles. These innovations aim to augment signal clarity while minimizing
interference.
2. Optimal Electrode Placement:
Precision in electrode positioning is under scrutiny to ensure accurate signal capture and to minimize
cross-talk or interference from adjacent muscle groups.
3. Enhanced Sensor Sensitivity:
Investigations into highly sensitive sensors capable of detecting minute muscle contractions are
ongoing. These sensors aim to offer more nuanced and detailed muscle activity data for improved
prosthetic control.
4. Advanced Signal Processing Techniques:
Engineers are developing sophisticated algorithms to refine EMG signal analysis. These techniques
aim to filter out noise, extract meaningful information, and decode muscle activation patterns
accurately.
5. User Training Initiatives:
Efforts are being directed toward creating user training programs to optimize the synchronization
between muscle signals and prosthetic device control. Such programs aim to enhance user
adaptability and control proficiency.
[6]
References
.
• Merletti, R., & Parker, P. A. (2004). Electromyography: Physiology, engineering, and non-
invasive applications. John Wiley & Sons.
• Farina, D., Merletti, R., & Enoka, R. M. (2014). The extraction of neural strategies from
the surface EMG. Journal of Applied Physiology, 117(11), 1215-1230.
• McGill, K. C., Lateva, Z. C., & Marateb, H. R. (2005). EMGLAB: An interactive EMG
decomposition program. Journal of Neuroscience Methods, 149(2), 121-133.
• Phinyomark, A., Phukpattaranont, P., & Limsakul, C. (2012). Feature reduction and
selection for EMG signal classification. Expert Systems with Applications, 39(8), 7420-
7431.
[7]
