AI Programs Help You Tackle Global Microplastics Challenges

Monash researchers have developed a new AI program to help scientists in the global battle against the tragedy of environmental microplastics. This study is published in the Journal of Hasardous Materials.

Despite making headlines in recent years, many scientists and policymakers still don’t know the scale of the question, including what kind of microplastics are there and where it ends.

Monash’s program uses sophisticated machine learning algorithms to analyze thousands of samples in fractions per second (a process that takes a few months for humans) and explain where and how they need to act. Get important understanding.

It’s not as easy as putting the sample under a microscope, as its appearance alone can be misleading.

Natural materials, such as small shells, often appear to be microplastics, for example.

Instead, new algorithms use the chemical components that make up these materials to accurately identify known microplastic types using data from a process called Fourier Transform Infrared spectroscopy (FTIR). “Signature” (complex numerical figures, thousands of characters) can be identified.

Importantly, this program is the first in the world to analyze a library of microplastic signatures. This is something researchers desperately need on the task of mammoths to address the problem.

The breakthrough was pioneered by lead researcher Frithjof Herb, a PhD from Monash University. Dr. Khay Fong, candidate, supervisor and senior lecturer at the Monash School of Chemistry.

“We are addressing key bottlenecks for progress in addressing the microplastics issue,” Herb said. “The process of analyzing samples is not only difficult and time-consuming, but we do it on a scale that is large enough to understand exactly what microplastics we’re dealing with, where we’re and where we’re. They couldn’t. They’ll end.

“This is a very important first step in finding ways to clean up these damaging microplastics and prevent them from entering the environmental waterway in the first place.”

In addition to shells, other natural fibers that are commonly mistaken for microplastics include algae, animal fur, or crustacean shells.

Herbs said that the evolution of human materials also complicates things, and the chemical composition of microplastics is constantly changing.

“Plastics are constantly changing, both in how they are made and how they break down in the environment. Traditional tools are struggling to keep up with these changes,” he said. said. “But our tools offer important benefits to scientists who need something that can be adapted quickly, which is important for analyzing ever-evolving data.

“We’re really proud of what we’ve achieved here. It works well with traditional laptops and focuses on sustainability and accessibility that we’ve sought through our small model.”