Projects

The publication describes a large EMG signal database including

• 43 subjects
  
• 3 days (over the span of 1 month)
  
• 16 classes of hand and finger gestures + rest (17 total)
  
• Wrist (2x6 channels) and forearm (2x8 channels)
  

In this project, upto 16 gestures were detected accurately in real time using wrist OR forearm EMG signals. The feature extracted from input signals were used to predict the gestures.

Biometrics are biological information or characteristics that can be used to identify and verify an individual’s identity. Conventional biometric traits such as fingerprints, facial recognition, and iris-scans are embedded in our daily lives in the form of authentication systems in smartphones, laptops, banks, etc. They provide improved security as compared to traditional password-based authentication systems or can be added as a second layer of security.

For the course project the task was to simulate an intra dural brain tumor removal surgery where the surgeon has to perform 4 stages of surgery:

1. incision on the skin layer
2. incision on the bone layer (involves milling and cutting)
3. incision on the dura matter
4. extraction of brain tumor

Using a gamification-based approach our objective was to simulate each stage and provide haptic feedback for each stage. Additionally, evaluation metrics for each layer were implemented as part of the gamification to record the learning curve of novice surgeons.

A working demo video is shown below.

For a course project it was proposed to build an augmented stylus and integrate it into a surgical training tool. The motivation for choosing such a topic is to provide low-cost (~200 CAD) haptic-based surgical training which can be used to train not only high school students but also professionals in the field. Currently, other high fidelity training tools have high graphical rendering. Our novelty lies with the cheaper cost and with the use of an augmented stylus for force sensing and haptic feedback.

The High Density Electromyography Signal processing tool is designed using MATLAB. The user interface allows to plot heat maps from multichannel EMG sensors. It also allows standard signal processing such as 1) filtering 2) change epoch size 3) extract Centre of Gravity (CoG) from the heat maps and 4) extract feature values at certain timestamps: amplitude, frequency, intensity, entropy.

The objective of the project was to explore force feedback designs using the Haply device. For this assignment, we were supposed to regulate the Haply movements using Proportional (P), Integrative (I) and Differentiate (D) control. Some fundamental questions regarding PID control were answered through this project.

The lab assignment was to learn the basics of how to use the Haply device. I first tried out the Wall and Fisica maze examples from the repository to play around and check out the features of the device.