Skip to main content

Dissertation Defense: Simon Ward, Electrical and Computer Engineering

Posted by on Monday, February 12, 2024 in News.

GraduateSchoolHeadshotEvent-75_Simon WardDISSERTATION DEFENSE

Simon J. Ward, Electrical and Computer Engineering
*under the direction of Sharon Weiss

“Improving Biosensor Performance using Machine Learning and Signal Processing

02.15.24  |  1:00PM CST  |  308 Featheringill Hall |
Zoom ID: 935 9293 1256

The field of biosensing, encompassing the detection of harmful biological molecules in medical diagnostic, food safety, and environmental monitoring contexts, plays a critical role in ensuring public safety and health. However, despite the promise of inexpensive, robust, portable, and user-friendly biosensors to revolutionize these application spaces, their performance often falls short of the requirements for clinical, physiological, or ecological relevance. This limited performance leaves many biosensing challenges unresolved or necessitates reliance on benchtop analytical instrumentation confined to centralized hospital labs. This work addresses the biosensor performance gap by showcasing the efficacy of a range of machine learning, statistical, and signal processing methodologies as tools to enhance biosensor performance. Notably, this research demonstrates the utility of deep learning-based time-series forecasting for reducing biosensor response time, the affordances of machine learning applied to biosensor-array data to obviate the need for capture agents, thereby increasing robustness and reducing cost, and the advantages of a novel signal processing algorithm leveraging Morlet wavelet filtering and Fourier analysis to reduce biosensor detection limits. These advances represent a step towards meeting the global need for robust, affordable, portable, easy-to-use, and rapidly responding biosensors with low detection limits in the fields of medical diagnostics, environmental monitoring, and food safety.


Tags: , ,