Motor Unit Conduction Velocity During Sustained Contraction Of The Vastus Med...
Muscle-Fatigue-Conf-Bhat
1. A Novel Approach to Detect Muscle
Fatigue During Isometric Exercises
PRAVEEN BHAT
INTELLIGRATED SOFTWARE
GRAND RAPIDS, MI
ADVISOR : DR. AJAY GUPTA
WIRELESS SENSORNETS (WISE) LABORATORY
DEPARTMENT OF COMPUTER SCIENCE
WESTERN MICHIGAN UNIVERSITY
KALAMAZOO, MI, 49008-5466, USA
2. Overview
• Previous Work
• Our Research Highlights
• Methodology
• System Components
• Results Analysis
• Conclusions
• Applications of Research
• Future Work
3. Few key terms
• Muscle fatigue
Loss of ability of a muscle or muscle group to produce a required or expected force
− Muscle fatigue is a result of prolonged or repetitive work
− The decrease in electrical stimulation of muscle leads to decline in force output
• Localized
Specific area of interest
• Isometric exercise
Muscle length do not change unlike in isotonic movements
4. Previous Work
• Non-invasive techniques are preferred
− Real-time monitoring, better comfort level
• Surface Electromyography(sEMG)
− Analysis of electrical activity of muscles captured
through surface electrodes
− Captured signal is called Electromyography(EMG)
signal
− Simple but reliable to observe muscle fatigue
− Used more often to evaluate lighter, repetitive work
Muscle Fatigue
Invasive
Blood Lactate Level
Blood Oxygen Level
pH of muscle
Needle Electrography Acoustic-Myography
Near-Infrared Spectroscopy
Surface Electromyography
Non-invasive
Sono-Myography
Mechano-myography
5. Domain and EMG Features
• Feature extraction is most important in EMG signal
analysis
• Frequency domain is preferred due to involvement of
dynamic muscle contractions
• Median and mean frequency are most frequently
examined features
• EMG features are extracted in time-frequency domain
in real time applications
Domain EMG features
Time domain
Mean Absolute Value(mAV)
Zero Crossings(ZC)
Slope Sign Changes(slopeSign)
Waveform Length(waveLen)
Root Mean Square(RMS)
Willison Amplitude(wAmp)
EMG Histogram(emgHist)
Frequency
domain
Median and Mean Frequency
Total Power
Mean Power
Peak Frequency
6. Summary of Previous Work
The spectral analysis of the surface Electromyography signal (sEMG) is
widely used to detect and quantify the electrical manifestations of
muscle fatigue.
Ref:
• Parmod Kumar, Anish Sebastian, and Chandrasekhar Potluri et al., “Spectral analysis of sEMG signals to investigate skeletal muscle fatigue”, CD-
ECC, pp. 47-52, Dec. 2011.
• A. Phinyomark, S. Thongpanja, H. Hu, and P. Phukpattaranont, “The usefulness of mean and median frequencies in electromyography analysis”,
pp.196-197, 2012.
7. So what was the opportunity ?
Previous research mainly focused on
− Various EMG analysis techniques and comparison between them
− Single parameter-based analysis rather than combining with other parameters
− Not directed towards developing wearable device
8. Proposed System
• Surface Electromyography with muscle vibration parameter.
• Simple but reliable
• Potential of developing wearable system
9. Methods
Our Algorithm
• Surface EMG (sEMG) signal is received periodically
• FFT is applied to transform sEMG signal into frequency
power spectrum.
• Calculate Median Frequency
• Rolling average method is applied to smooth out periodic
fluctuations of sEMG signal.
• Calculate vibration speed in parallel
• Muscle fatigue is indicated when median frequency value
decreases while vibration speed progressively increases.
0
1
2
3
4
5
6
1 2 3 4 5 6 7 8 9 10
Time(s)
Median Frequency Vibration
Imaginary graph of Median frequency
vs. vibration during isometric exercises
10. System components
• BITalino Board
− Specially designed for biosignal exploration
− Sampling rate 1, 10, 100, and 1000 Hz
− Consists of sensors EMG, ECG, EDA, and Accelerometer
• Electrodes
• Mobile device
BITalino Board
Fatigue Detection System
Accessories
11. Experimental Setup and Procedure
• Three surface electrodes were placed on subject’s
biceps brachii.
• A weight of 5 pound was handed to the subject to
hold parallel to horizontal axis.
• Subjects were asked to hold the weight until they
feel exhaustion.
12. Experimental Highlights
• 30+ participants of age ranging from 21 to 54 years, both male and
female
• Captured 3 readings from each participant with sufficient rest time
• Consent form signed by participants before participation (followed
HSRIB procedure)
16. Vibration values across participants
S1
0
200
400
600
800
1000
1
9
17
25
33
41
49
57
65
73
81
89
97
105
113
121
129
137
145
153
161
169
177
185
193
201
209
217
225
233
241
249
257
Speed
Vibration Number
A participant’s steady state for long duration
Initial fluctuations
17. Median frequency and vibration combined
0
200
400
600
800
1000
1200
1400
1600
1800
0
5
10
15
20
25
30
35
40
45
Vibrationspeed
Medianfrequency
Decreasing trend
A participant’s median frequency not changed !
18. Fatigue Detection for a Participant
-6
-4
-2
0
2
4
6
8
-150
-100
-50
0
50
100
150
1
3
5
7
9
11
13
15
17
19
21
23
25
27
29
31
33
35
37
39
41
43
45
47
49
51
53
Averagerateofchange
Time(s)
Vibration Median Frequency
Fatigue detection point
Fatigue detection point
19. Conclusions
• Observed progressive vibration increase while median frequency is
decreasing
• Vibration re-enforces the fatigue detection thereby improving the
accuracy and reliability of a model
• No significant difference between male and female participants
• Practical approach to the development of a wearable device
20. Applications of Research
• Ergonomics
• Sports
• Medical science
• Health management system and many more !
21. Future work
• Study of muscle fatigue on different “locality” of muscles
• Further analysis of anomalies that were encountered
• Combine health profile to make the system more intelligent
• What about vibration data during dynamic contractions ?
• Comparison with other models
23. References
• J.D. Rouillon, R. Candau, La fatigue périphériq ue: sites subcellulaires et mécanismes
biologiques, Sci. Sports , vol. 15, no. 5, pp 234-241, Oct. 2000.
• John M. Heasman, Timothy R. D. Scott, and Veronica A. Vare et al., "Detection of fatigue
in the isometric electrical activation of paralyzed hand muscles of persons with
tetraplegia," IEEE Transaction on Rehabilitation Engineering, vol.8, no. 3, Sept. 2000.
• Daniel T. H. Lai, Rezaul K. Begg, and Marimuthu Palaniswami, “Computational
Intelligence in Electromyography Analysis – A Perspective on Current Applications and
Future Challenges,” IEEE Transl. on information technology in biomedicine, vol. 13, no. 5,
Sept. 2009.
• Kenichi Ito and Yu Hotta, “Surface electromyogram–based detection of muscle fatigue
during cyclic dynamic contraction under blood flow restriction”, 36th Annual
International Conference on Engineering in Medicine and Biology Society (EMBC), 2014.
• Rubana H. Chowdhury, Mamun B. I. Reaz, and Mohd Alauddin Bin Mohd Ali et
al."Surface electromyography signal processing and classi_cation Techniques," Sensors, 13,
12431-12466, 2013.
24. References
• D.T.MacIsaac, P.A.Parker, and K.B.Englehart, “A Novel Approach to Localized Muscle
Fatigue Assessment”,vol.3, pp. 2487-2490, Sept. 2003.
• Basmajian, J.V. and C.J. De Luca,”Muscle alive: their function revealed by
electromyography”, 5a ed. Baltimore, Williams e Wiikins, pp. 501-561, 1985.
• Mohamed R. Al-Mulla , Francisco Sepulveda, and Martin Colley, “A Review of Non-Invasive
Techniques to Detect and Predict Localised Muscle Fatigue”, 11, 3545-3594, Sensors 2011.
• Parmod Kumar, Anish Sebastian, and Chandrasekhar Potluri et al., “Spectral analysis of sEMG
signals to investigate skeletal muscle fatigue”, CD-ECC, pp. 47-52, Dec. 2011.
• R. Merletti, L. R. Lo Conte, and C. Orizio, “Indices of muscle fatigue,” Journal of Electromyography
and Kinesiology, 1(1), p p. 20-33, 1991.
• F.B. Stulen and C.J. De Luca, “Frequency parameters of the myoelectric signal as a measure of
muscle conduction velocity”, IEEE transl. vol. 7, July 1981.
• A. Phinyomark, S. Thongpanja, H. Hu, and P. Phukpattaranont, “The usefulness of mean and
median frequencies in electromyography analysis”, pp.196-197, 2012.
Notas del editor
Near-Infrared Spectroscopy: Method to determine the level of muscle blood volume(BV) and oxygenation of muscle. It uses some advanced optical instrumentation to measure hemoglobin values.
Mechano-myography: mechanical characteristics of a muscle. Helps to investigate motor unit activity and get insights of intrinsic mechanical activity.
Acoustic-myography: Acoustic myography is the recording of sounds produced by contracting muscle.
These sounds become louder with increasing force of contraction
Sono-Myography: using ultra-sound signal to detect the dynamic muscle thickness changes in skeletal muscle