The Classification of EMG Signals using Machine Learning for the Construction of a Silent Speech Interface
Research Publications (click to Expand)
The Young Researcher (Peer-Reviewed Research Journal):
Description: I was one of 20 students in the nation to be selected for publication in "The Young Researcher" - a peer-reviewed journal dedicated to publishing the best original research from secondary school students.
HyperLink: The Young Researcher Publication
Description: I was one of 2 students in the nation selected by the College Board, whose research work was published as a perfect scoring exemplar model for the AP Research course.
HyperLink: College Board Publication
The STEMY Innovation Journal:
Description: The STEMY nonprofit creates its own research journal. This journal presents ambitious research projects for a younger audience so that students can be inspired and more involved with STEM.
HyperLink: STEMY Innovation Journal Publication
With 7.5 million people unable to speak due to various conditions, patients are forced to use cumbersome/inefficient devices such as eye/cheek trackers. In this study, a speech aid known as a Silent-Speech-Interface(SSI) was created. This device could be used by patients with speech disorders to communicate letters in the English-alphabet voicelessly, merely by articulating words or sentences in the mouth without producing any sounds. The SSI records EMG signals from the speech system which are then classified into speech in real-time using a trained Machine Learning(ML) model. It was found that the SVM-ML algorithm yielded the highest SSI accuracy of 90.1% Overall, this study involves the creation of a device that measures biomedical signals and translates them into speech accurately using an ML algorithm. This study’s findings could improve the accuracy of future SSIs by identifying the most accurate algorithms for use in an SSI.
Automatic Portable Phoropter and Auto Focus Glasses using an Integrated Variable Focal Liquid Lens
A third of the world’s population has poor vision and can’t obtain visual aids, negatively impacting economies and societies. People also use multifocal eyeglasses instead of using multiple glasses with various prescriptions. Although multifocal lenses combat the need for multiple glasses, it confines vision by only providing users a narrow field to view objects. These glasses cause eye strain/fatigue and chronic neck problems/pain. This project aims to create a device that can integrate a voltage-controlled liquid lens to change lens curvature to mimic phoropters and multifocal visual aids without their associated problems. Data obtained on the “percent increase in visual acuity” in relation to participants' original visual acuity show that the average increase using auto focal glasses (58.06%) was significantly higher than the average increase using current visual aids (41.33%). This difference was determined to be significant showing with 95% confidence that the device was able to improve vision in comparison to commonly used visual aids. This device improves vision as it allows for finer adjustments to be made by the user to focus light on the optic nerve. The device was also effective in improving vision due to the blocking of stray light caused by the reduced aperture of the lens which increases visual acuity. In conclusion, this auto focal visual aid improved a user’s vision significantly more than the user’s current visual aids as these auto focal glasses can adjust and adapt to the user’s current visual needs.
The Effect of Light Tracking and Tower Based Solar Energy Systems on Photovoltaic Energy Generation
Solar energy is the cleanest and most abundantly available energy source. Solar technologies can harness this energy to produce electricity. My project aims to create an economic and viable solar energy harvester which would be adaptive to weather conditions and would be cost-effective. This project also intends to compare the efficiency of Tower Based Solar Systems to Light Tracking Solar Systems.
Hypothesis: If Solar panels follow the sun using a Solar Tracking System, then the panels would generate more photovoltaic energy than the Tower based system which has mirrors positioned at various angles to reflect sunlight onto the solar panel throughout the day.
The Solar Tower was built using a PVC pipe which was mounted between the 3-D printed prism and base. The solar Tracker was made by putting two servos into a 3-D printed coaxial holder.
The photovoltaic energy generated by all the harvesting methods followed the same patterns of drops and increases. Amongst all the models, the average photovoltaic energy generated by the Solar tracker was the highest and stationary solar panels had the lowest.
The T-values in show that both the Solar Tracker and the Solar Tower harvested significantly higher amounts of energy when compared to the Stationary solar panels. The data obtained, thus suggests that both the Solar tracker and the Solar Tower are much more efficient methods to harvest solar energy