New method to detect microplastics is using AI

A team out of the University of Waterloo developed an AI tool called PlasticNet, it helps researchers quickly analyze a large number of particles close to 50 per cent faster and 20 per cent more accurate.

They believe it’s the first time anyone’s ever used artificial intelligence (AI) to detect microplastics in wastewater samples.

With the growing concern about microplastics, Dr. Wayne Parker, a professor in Waterloo’s department of civil and environmental engineering wanted to find a way they could find microplastics faster, so that people can get rid of them faster.

“We wanted to make use of advanced micro scoping to generate essentially a fingerprint of microplastics that we saw in wastewaters and then we wanted to use artificial intelligence to interpret those fingerprints in a meaningful way,” said Parker. “So, we could more accurately identify the plastics and differentiate them from other kinds of particles that are often present in wastewaters.”

Parker teaches wastewater and waste management. He approached his colleague Dr. Alexander Wong, a professor in Waterloo’s department of systems design engineering and the Canada Research Chair in AI and medical imaging for help creating PlasticNet.

“People in the past have used microscopes to visually identify microplastics so there was a lot of human error there,” said Parker.

The team out of the university also consists of PhD students like Frank Zhu, who is helping look at what AI is finding from analyzing the wastewater samples they use.

When they started the project, they realized a lot of methods for identifying microplastics in these types of wastewaters were not working well and slow to use.

“We found that some of our methods were faster than some of the previous methods that have been employed for this purpose and its also more accurate.”

Parker said they were able to train the artificial intelligence to be able to recognize from a broad range of plastics that are out there, which in turn, made the new method became more accurate.

“Some of these tools that we’ve developed in the first phase of the work now to look at applications to better understand the movement of microplastics through wastewater treatment plants. So that we can better predict in the future where microplastics are going and to evaluate any risks that go along with those microplastics,” said Parker.

Parker added that as awareness about microplastics grows, they are trying to figure out any risk that might come out as a result of those microplastics. The team out of the University of Waterloo is trying to understand what needs to happen next to deal with microplastics in wastewater samples.

“Microplastics are everywhere now.”