UWaterloo uses AI to improve microplastics classification process 

0
microplastics-research-image
The research team said it tested its advanced imaging identification system on microplastics isolated from a local wastewater treatment plant and found great success in advancing the fight against the potential threats posed by microplastics to aquatic and human health. Photo Credit: Pcess609, stock.adobe.com

The University of Waterloo (UWaterloo) has developed an AI tool that enables researchers to rapidly analyze large numbers of microplastic particles some 50% faster and with 20% more accuracy than previous methods.

The AI deep-learning tool, called PlasticNet, addresses the challenge that microplastics come in wide varieties due to the presence of manufacturing additives and fillers that can blur the “fingerprints” in a lab setting. For humans, that process of identifying microplastics from organic material was not only slow, but prone to error, said UWaterloo researchers. 

Water Institute member Dr. Wayne Parker and his team approached Dr. Alexander Wong, a professor in Waterloo’s Department of Systems Design Engineering and the Canada Research Chair in Artificial Intelligence and Medical Imaging, for assistance in developing a tool to improve the microplastics identification and classification process.

“We trained it on data from existing literature sources and our own generated images to understand the varied make-up of microplastics and spot the differences quickly and correctly — regardless of the fingerprint quality,” explained Wong in an announcement from the university.

Subscribe to our Newsletter!

The latest environmental engineering news direct to your inbox. You can unsubscribe at any time.

According to the team’s new research paper, PlasticNet successfully classified 11 types of common plastic with an accuracy higher than 95% after it was trained with more than 8,000 spectra of virgin plastic.

The images analyzed by PlasticNet were generated by focal plane array-based spectroscopy combined with traditional light microscopy.

The research team said it tested its advanced imaging identification system on microplastics isolated from a local wastewater treatment plant and found great success in advancing the fight against the potential threats posed by microplastics to aquatic and human health.

More conventionally, the UWaterloo researchers said microplastics are identified by using a library search strategy. Users can search for a correlation coefficient between an unknown spectrum and one or few standard spectra in a reference library “calculated to determine which plastic type most closely matches the unknown spectrum.”

UWaterloo researchers added that: “There is a growing interest in leveraging machine learning models for spectral classification and [microplastics] detection.”

Dr. Parker at the university’s Water Institute noted that microplastics are hydrophobic materials that can soak up other chemicals. 

“Science is still evolving in terms of how bad the problem is, but it’s theoretically possible that microplastics are enhancing the accumulation of toxic substances in the food chain,” Parker said in a statement from the university. 

No posts to display

LEAVE A REPLY

Please enter your comment!
Please enter your name here