Improving technology for identifying environmental microplastics with machine learning

The average human consumes a credit card worth of plastic every week as a result of environmental microplastics. The tools and technology that are currently used to analyze chemical compound structures to identify polymer types in microplastics research are not well-calibrated for field-specific use. Raman spectroscopy data from microplastics samples is imperfect. Furthermore, plastics that have been weathered by environmental factors offer even less analytic certainty. Various environmental factors further skew the spectroscopy data. Machine learning tools and techniques can allow us to better calibrate the research tools for certainty in microplastics analysis.
This research study will not only improve our knowledge of microplastics spectroscopy data broadly, but also break new ground into understanding chemical compounds of plastics that have been weathered by various environmental processes. The significance of this project is to strive for a measurably improved predictive capacity of Raman spectroscopy data to help classify polymer types through an applied machine learning process.

Faculty Supervisor:

Sheela Ramanna

Student:

Partner:

Compound Connect

Discipline:

Computer science

Sector:

Professional, scientific and technical services

University:

University of Winnipeg

Program: