UC Berkeley's Computational Microscope Captures 25.2 Billion Pixels Per Second Across a Wide Field of View
A UC Berkeley team built a 48-sensor computational microscope that hits 3-micron resolution across 5 square centimeters at 120 frames per second.
Overview
Researchers at the University of California, Berkeley have built a computational microscope that captures 25.2 billion pixels per second, a rate the team says overcomes the long-standing trade-off between speed, field of view, and resolution in optical imaging, according to UC Berkeley’s College of Engineering. The system reached 3-micron resolution across 5 square centimeters at 120 frames per second, as reported by Phys.org.
The work is led by Laura Waller, a professor of electrical engineering and computer sciences at Berkeley, according to UC Berkeley. “This is really a breakthrough in the field of computational microscopy,” Waller said, per UC Berkeley. “With our microscope, we were able to achieve 3-micron resolution across 5 square centimeters at 120 frames per second, which isn’t possible with traditional ways. It’s the largest space-bandwidth time product of any practical microscope we know of,” she said, according to UC Berkeley.
What We Know
The microscope combines an array of 48 camera sensors, housed on a credit-card-sized circuit board, with an engineered diffractive optical element and a computational reconstruction algorithm, UC Berkeley reported. Waller described the optical component in detail: “Specifically, we fabricated a custom-designed phase mask — a glass plate that diffracts light in the microscope — so that light that would normally fall between the sensors is redirected onto them instead,” she said. A reconstruction algorithm then fills in the remaining data gaps created by the physical spacing between sensors, according to UC Berkeley.
The team demonstrated the system by imaging dozens of freely moving C. elegans nematodes for 15 seconds at 120 frames per second, UC Berkeley reported. Kevin C. Zhou, the study’s lead author and a former Berkeley postdoctoral researcher now an assistant professor at the University of Michigan, said the microscope’s combination of resolution, field of view, and speed “allowed us to track the freely moving C. elegans, as well as some structures within them,” according to UC Berkeley. “From the video reconstruction, for example, we were able to track individual worms and perform functional imaging of their rapid pharyngeal pumping, a part of their feeding behavior,” Zhou said. Phys.org separately reported that the system eliminated the need for the manual calibration typically required in large-scale microscope setups, according to Phys.org.
Chaoying Gu, a Berkeley Ph.D. student in electrical engineering and computer sciences who contributed to the work, tied that calibration-free design to future scaling: “I think calibration-free capabilities like our method will prove to be one of the key ingredients for scaling up imaging systems in the future,” Gu said.
The research, published in the journal Nature Photonics, involved collaborators beyond Berkeley, including researchers at UC San Francisco, the University of Utah, UC San Diego, Duke University, and the optics company Ramona Optics, according to UC Berkeley. The project drew funding from the Office of Naval Research, the Air Force Office of Scientific Research, the National Institutes of Health, the Japan Society for the Promotion of Science, the Schmidt Science Fellows program in partnership with the Rhodes Trust, and the Chan Zuckerberg Biohub San Francisco, UC Berkeley reported.
Waller framed the broader goal behind the project as addressing a persistent gap in the field: “There’s a lot of progress being made, and still to be made, on very large-scale video microscopy. Our study shows that computational imaging — the joint design of hardware and software — has a lot to offer in scaling up to large data acquisition,” she said. She added that the approach “could be used to image many live organisms simultaneously and to monitor samples over time,” according to UC Berkeley.
What We Don’t Know
Neither UC Berkeley’s release nor Phys.org’s coverage disclosed a timeline for commercializing or deploying the system beyond the lab, nor did they specify the cost of building the 48-sensor array. The full technical details of the reconstruction algorithm and phase-mask design are published in the Nature Photonics paper, which was not independently accessible for this report.