Signal Processing and Applied Machine Learning for the Detection of Nucleic Acids

The rapid detection of microorganisms using DNA-based solutions has positive, far-reaching consequences on society for curing diseases. Unfortunately, current DNA-based methods are expensive, non-portable and require a technically trained operator. These challenges can be tackled from three different angles: (i) improving and developing novel detection chemistries; (ii) introducing new engineering solutions (such as semiconductor-based chemical imaging devices); and (iii) developing signal processing and machine learning algorithms in order to extract as much information from the data as possible.

The primary aim of this project focuses on the latter: addressing the role of signal processing, machine learning and data science for enhancing DNA-based solutions for problems in healthcare. Examples of such problems include: detecting pathogens, quantifying how much of a pathogen exists, and detecting multiple pathogens in a single chemical reaction. The goal is to touch on these issues using data obtained from: standard instruments, state-of-the-art instruments, and then tackling new data obtained from semiconductor-based chemical imaging devices. Although the novelty of this work lies in developing new algorithms for extracting information from data (obtained from DNA-based experiments), the benefits of this project inherently extend to those enhancing detection chemistries and developing new engineering solutions.

Research Areas
Artificial intelligence technologies, Digital signal processing and Mathematical Biology