1. Real – Time Locomotion Classification using Transient Surface EMG signals
Sarthak Pati1, Deepak Joshi2, Ashutosh Mishra2 and Sneh Anand2
1 – Dept. Of Biomedical Engineering, Manipal University
2 – Center for Biomedical Engineering, IIT – Delhi

2. Contents Introduction to EMG and its acquisition Importance of EMG
Pre – Processing of EMG signals
Features under consideration
Classifier design

3. Fig 1 : EMG Signal of Healthy Subject
What is EMG ?
It is a signal used to evaluate the electrical activity produced by skeletal muscles.
Fig 1 : EMG Signal of Healthy Subject

4. Block Diagram EMG Acquisition Signal Processing Feature Extraction
Classification
Using Basic Surface Electrodes
Band Pass Filtering
Time – Domain Features
Linear Discriminant Analysis

5. EMG Surface Electrodes
Fig 2 : EMG Surface Electrodes
Image Courtesy : Orthotics and Prosthetics Lab, BME Unit, AIIMS

6. Electrode Placement Fig 3 : Electrode Placement Diagram
Image Courtesy : Orthotics and Prosthetics Lab, BME Unit, AIIMS

7. Importance of EMG Diagnosis of Neuro - Muscular Disorders
Motor Control Disorders
Prosthetic Control
Sensing of Isometric Motor Activity (motion–less gestures)
Flight control (Human Senses Group, NASA)
Machine–Human Interfacing (Advanced Robotics, MIT)

8. Why EMG for this study ? Relatively easy to acquire and process
If properly utilised, gives good accuracy for control systems
High sensitivity
Single Muscle Recording Possible
Access to Deep Musculature
Little cross – talk concern

9. EMG – Signal Processing
Fig 4 : Frequency Response of Band Pass Filter

10. Feature Selection Criteria : Computational Efficiency
High separability with respect to locomotion modes

11. Classifier Design Obtaining LDA Transformation Matrix T
Each Locomotion Mode mapped to a single dimension data set using T
Threshold – based approach for classification

12. Fig 5 : LDA classification between all the four locomotion modes
Results
Fig 5 : LDA classification between all the four locomotion modes

13. Fig 6 : LDA classification between FW and SW
Continued…
Fig 6 : LDA classification between FW and SW

14. References
Deepak Joshi, Sneh Anand - Study of circular cross correlation and phase lag to estimate knee angle: an application to prosthesis; Int. J. Biomechatronics and Biomedical Robotics [in press]
Hargrove L. J., Huang H., Schultz A. E., Lock B. A., Lipschutz R., Kuiken T. A. - Toward the Development of a Neural Interface for Lower Limb Prosthesis Control; Delsys Prize Winner
Parker P., Englehart K., Hudgins B. - Myoelectric signal processing for control of powered limb prostheses; Journal of Electromyography and Kinesiology
Englehart K., Hudgins B. - A Robust, Real-Time Control Scheme for Multifunction Myoelectric Control; IEEE Transactions on Biomedical Engineering, Vol.50, No.7
Chan F.H.Y., Yang Y.S., Lam F.K., Zhang Y.T., Parker P.A. - Fuzzy EMG Classification for Prosthesis Control; IEEE Transactions on Rehabilitation Engineering, Vol.8, No.3
Englehart K., Hudgins B., Parker P., Maryhelen S. - Time-Frequency Representation for Classification of The Transient Myoelectric Signal; 20th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Vol. 20, No 5
Phinyomark A., Limsakul C., Phukpattaranont P. - A Novel Feature Extraction for Robust EMG Pattern Recognition; Journal of Computing, Vol 1, Issue 1, ISSN:

15. Thank You
Any questions…?