Machine Learning Can Speed Up Microplastic Counting
Researchers at the University of Toronto’s Faculty of Applied Sciences & Engineering have proposed new methods that will use machine learning to make counting and classifying microplastics easier, faster, and more cost-effective.
Elodie Passeport, an associate professor in the departments of civil and mineral engi-neering and chemical engineering; Shuyao Tan, a Ph.D. student in chemical engineering; and Joshua Taylor, an associate professor in the department of electrical and computer engineering, worked together on the investigation, which was published in ACS ES&T Water. Using an algorithm that will not introduce additional error or variance allows only a small fraction of samples to be manually processed so the rest will be predicted. This project is the first open-source dataset for microplastics image segmentation.
High School Students design a Bottle that Makes Seawater Potable
Four high school students have designed a desalinating water bottle that could make seawater potable. The design was inspired by the drinking straws hikers use to purify water, considering if it was possible to make a bottle that could do the same for seawater.
The students (Laurel Hudson, Gracie Cornish, Kathleen Troy, and Maia Vollen) met in Virginia Tech’s C-Tech2 program. The program challenged the four students to “reinvent the wheel,” so they focused their assignment on the ongoing global water crisis.
The group has published their findings in the journal Soft Matter, and while the design is theoretical, it could prove to provide a way for communities in which drinking water is scarce to be able to rely on seawater to fulfill their needs.
Researchers Develop Techniques to Test Outside of a Lab
Joe Sinfield, professor of civil engineering, and researchers in his lab at Purdue University’s Lyles School of Civil Engineering have created a suite of patented and patent-pending technologies using Raman spectroscopy—a chemical analysis technique that uses light to assess the chemical composition of materials.
This technology allows for chemical analyses on a wide array of sensitive compounds outside of the laboratory, which results in a low-to-moderate cost and little need for expertise of sample preparation. Sinfield’s new inventions include addressing interference from fluorescence, addressing problems created by objects in samples that are not usually of interest to researchers, enhancing the Raman system sensitivity to enable chemical analyses in challenging situations, and enabling spatially dispersed analyses.
From Seawater to Drinking Water with the Push of a Button
Researchers at MIT have developed a portable desalination unit that weighs less than 10 kilograms. This new device is suitcase-sized, requires less power to operate than a cell phone charger, and is packaged in a user-friendly device that runs with the push of a button. The device will automatically generate drinking water that exceeds World Health Organization quality standards.
The research has been published in Environmental Science and Technology. Senior author Jongyoon Han is a professor of electrical engineering and computer science and of biological engineering at MIT and a member of the Research Laboratory of Electronics (RLE).
Junghyo Yoon, a research scientist in RLE; Hyukjin J. Kwon, a former postdoc; SungKu Kang, a postdoc at Northeastern University; and Eric Brack of the U.S. Army Combat Capabilities Development Command (DEVCOM), are also authors.
This device, unlike others of its kind that require water to pass through filters, uses electrical power to remove particles from drinking water, which eliminates the need for replacement filters and reduces long-term maintenance requirements. The design and function of this device could allow for it to be deployed in remote and severely resource-limited areas.