UCLA’s imaging flow cytometer is powered by deep learning. Credit: Zoltan Gorocs, Miu Tamamitsu,Vittorio Bianco, Patrick Wolf,Shounak Roy, Koyoshi Shindo, Kyrollos Yanny, Yichen Wu, Hatice Ceylan Koydemir, Yair Rivenson & Aydogan Ozcan. Light: Science & Applications (2018) 7:66. DOI 10.1038/s41377-018-0067. Click image to enlarge.

UCLA researchers have developed an inexpensive and portable device that can analyze water samples immediately, which would provide marine biologists with real-time insight about the possibility that algal blooms could occur in the area they’re testing. That, in turn, would allow officials who manage coastal areas to make better, faster decisions about, for example, closing beaches and shellfish beds before algal blooms cause serious damage.

The researchers created a new flow cytometer — which detects and measures the physical and chemical characteristics of tiny objects within a sample — based on holographic imaging and artificial intelligence. It can quickly analyze the composition of various plankton species within a matter of seconds.

The research, which was published online by Light: Science & Applications and will appear in the journal’s print edition, was led by Aydogan Ozcan, the UCLA Chancellor’s Professor of Electrical and Computer Engineering and associate director of the California NanoSystems Institute at UCLA.

Scientists have generally tried to understand algal blooms through manual sampling and traditional light microscopy used to create high-resolution maps that show the phytoplankton composition over extended periods of time. To build those maps, technicians have to collect water samples by hand using plankton nets, and then bring them to a lab for analysis. The process is also challenging because the concentration and composition of algae in a given body of water can change quickly — even in the time it takes to analyze samples.

The compact and lightweight device created by Dr. Ozcan and his colleagues speeds up the entire process and, because it does not use lenses or other optical components, it performs the testing at a much lower cost; it can be assembled from parts costing less than $2,500. It images algae samples — and is capable of scanning a wide range of other substances, too – using holography and artificial intelligence.

One challenge the researchers had to overcome was ensuring the device would have enough light to create well-lit, high-speed images without motion blur.

“It’s similar to taking a picture of a Formula 1 race car,” Dr. Ozcan said. “The cameraman needs a very short exposure to avoid motion blur. In our case, that means using a very bright, pulsed light source with a pulse length about one-thousandth the duration of the blink of an eye.”

To test the device, the scientists measured ocean samples along the Los Angeles coastline and obtaining images of its phytoplankton composition. They also measured the concentration of a potentially toxic alga called Pseudo-nitzschia along six public beaches in the region. The UCLA researchers’ measurements were comparable to those in a recent study by the California Department of Public Health’s Marine Biotoxin Monitoring Program.

Zoltán Gӧrӧcs, a UCLA postdoctoral scholar and the study’s first author, said the researchers are in the process of discussing their new device with marine biologists to determine where it would be most useful. “Our device can be adapted to look at larger organisms with a higher throughput or look at smaller ones with a better image quality while sacrificing some of the throughput,” Dr. Gӧrӧcs said.

Other co-authors of the paper are Miu Tamamitsu, Vittorio Bianco, Patrick Wolf, Shounak Roy, Koyoshi Shindo, Kyrollos Yanny, Yichen Wu, Hatice Ceylan Koydemir and Yair Rivenson.

The research was funded by the U.S. Army Research Office.

Source: Changchun Institute of Optics, Fine Mechanics and Physics