WOBURN, MA—Cambridge Research & Instrumentation (CRi) develops and uses imaging technology that makes it possible to see the otherwise unseeable. Since it was founded in 1986, CRi has produced advanced imaging technologies intended to complement scientists and technologists working in basic research, drug development, customized medicine, and other pursuits in life science and beyond. Its optical techniques enable imaging at size scales ranging from single cells to live animals.
CRi serves a wide range of customers. “They include lots of academics, as well as government institutions such as the National Institutes of Health and other national laboratories. And we sell a lot of devices into industry, particularly pharma and biotechnology,” says VP, CTO, and co-founder Clifford Hoyt. “We serve anybody interested in getting more information out of complicated samples; we have a way of teasing out the information.”
The company focuses mainly on applications in life science. Its fluorescence microscopy systems permit scientists to differentiate multiple protein and pathway configurations in tissue sections. Its in vivo molecular imaging systems provide key information about proteins and RNA species in intact small animals. And its adaptation of technology invented at the Marine Biological Laboratory in Woods Hole, MA has led to a microscope that enables nondestructive assessment of the health of ova, leading to advances in infertility medicine and stem cell research.
Dead Sea Scrolls
However, the technology has produced results in areas outside the life sciences. Last summer, for example, a California scientist took one of the company’s Nuance multispectral imaging systems to the Israel Museum to create near-infrared shots of the Dead Sea Scrolls. The goal: to reveal text that is invisible to the human eye on the Scrolls’ fragmentary, discolored parchment and papyrus. A scientific team will use the data gathered in this way as part of a new conservation program that will monitor the scrolls for changes.
CRi’s development of Nuance and other offerings started two decades ago, when it received a patent for liquid-crystal servo control and liquid-crystal tunable filters. In the next few years, the company continued to develop its tunable filter technology. “That was one of our core technologies, along with a continuously tunable impasse filter,” Hoyt recalls. Then, as the company sold those filters, further vistas became evident. “We learned a lot about the applications of the tunable filters, some of which we though had significant market promise” Hoyt says. “We saw customers having to undertake an engineering project to fit our filters into their systems. Customers also had a problem with the software; the solutions weren’t always scalable or effective.” In response, he continues, “we developed powerful algorithms for taking multispectral images and ‘spectrally unmixing them’ into the components that overlap in the multispectral images.”
Each pixel in a multispectral image contains an entire spectrum of the light in its location—between 5 and 20 data points rather than the typical 3 in a red-green-blue image. So the process allows significantly more quantitative and sensitive measurement of subtle differences in spectral emissions, and hence increased sensitivity when the images are unmixed. The entire process, the company’s website states, “represents a dramatic development in fluorescence microscopy. Not only does it allow for the use and detection of many molecular labels simultaneously, but by removing the confounding effects of autofluorescence, it can provide up to a 300-fold increase in sensitivity, often turning seemingly negative results positive.”
One of the top ten
CRi’s technology gained recognition last November when the Cleveland Clinic ranked it as one of the year’s top ten medical innovations. “When we’re talking about tumors, we’re talking about what information we can gain about that tumor to guide and direct therapy, prognosis, and diagnostics,” said Jennifer Hunt, the clinic’s head of surgical pathology. “We believe this technology to be a game changer. Being able to analyze multiple markers in a single cell to understand the behavior of signaling pathways will significantly aid in disease diagnosis and therapy development.”
Most users experience the technology in the form of the Nuance camera, which CRi developed after customers reported difficulty integrating filters into their imaging system. “We solved a couple of integration problems by developing the camera. It involved acquiring the images in a synchronized way with the filter tuning and avoiding degrading the image,” Hoyt explains. “To the user, it’s a simple camera. But is can do things no other camera can.”
Another recent addition to the company’s product line, the Maestro system, permits scientists to produce images on a far larger scale. “We found that many of our Nuance customers were imaging whole small animals in vivo,” Hoyt says. “They had integration issues like illumination control and light-tight enclosure. The Maestro is essentially a specialized Nuance camera in a big box with thoughtfully designed optics and automation.” This system’s main advantage is in removing the autofluorescence that effectively represents background noise. “Customers have found that you improve your signal-to-background ratio by factors up to 300. That’s useful in measuring emission levels inside living animals,” Hoyt says. “It will measure any animal that fits into an 8-inch by 8-inch field of view. And it has an extended range from blue to near-infrared—950 nm, where absorption of small animal tissue is much lower, so that you can see deeper in the animals.”
New functionality
What does the company plan next? “We are now adding functionality to provide customers with more of the solution,” Hoyt says. “Our latest innovation is on the image analysis side. Our systems provide exquisite imagery, but most customers want data.” To that end, the company has just released a new software package based on machine learning. “As the operator, you need to recognize what you are looking for with your eyes and click the mouse onto it; the system will find the objects you care about and extract the spectrally unmixed items in a form that you can use for data mining and data analysis packages,” Hoyt explains. “It works very fast to find a classifier or an algorithm. And it can work on a standard laptop. We’re using it in exciting applications in cancer research.”
Peter Gwynne | Freelance writer
Peter Gwynne is a freelance writer based in Massachusetts; e-mail: [email protected].