An endoscope equipped with an infrared laser and a tiny mirror might one day help physicians diagnose early signs of cancer and other diseases and aid in surgery. A researcher at the University of Florida has designed a prototype device that captures images up to two millimeters beneath the surface of tissues, providing high-resolution, three-dimensional images at video-rate speeds.

In typical endoscopy, doctors thread a long, thin, camera-equipped fiber through a patient’s airway or gastrointestinal tract to search out abnormalities. The images, displayed on a monitor in real time, can reveal signs of infection, internal bleeding, ulcers, and tumors on tissue surfaces. But today’s endoscopes only show a superficial picture–they don’t reveal what’s going on under the surface, such as early tumor development.

“Eighty-five percent of cancers originate from the epithelium, which is about two millimeters deep,” says Huikai Xie, associate professor of electrical and computer engineering and director of the Biophotonics and Microsystems Laboratory. In addition to its potential for detecting early signs of cancer, he says, the scope might prove useful as a surgical tool, helping surgeons determine how deep a tumor is embedded in tissue. “If you need to remove the tumor, the surgeons have a hard time determining when to stop. With a real-time, high-resolution tool, they will be sure.”

John Saltzman, a gastroenterologist and director of endoscopy at Brigham and Women’s Hospital, says such a technique would help identify early signs of cancer, particularly in the esophagus. In a condition called Barett’s esophagus, for example, cells lining the esophagus undergo a change that increases the risk of cancer, says Saltzman, who is not involved in the research. “This technology would be an advantage for us to detect such abnormalities.”

Instead of a tiny camera at the tip, Xie’s endoscope is equipped with an infrared scanner and a tiny mirror, which scans tissue layer by layer to provide a three-dimensional image with microscopic resolution. The technique is based on a method called optical coherence tomography (OCT)–as a laser beams through the arm of an OCT scope, it hits tissue, and reflects some light back, while the rest scatters. Different tissues, such as cancer versus normal tissue, reflect light differently. An interferometer measures the reflected light and subtracts the scattered light. Altering the length of the arm alters the depth at which light is directly reflected back, producing images of different layers, which together form a three-dimensional image. The method is similar to ultrasound technology, and is often called “optical ultrasound.”

Today, OCT is used in optometry to image the retina for signs of glaucoma and macular degeneration. That technology, used to scan outside the body, involves bulky equipment requiring a lot of power. Only recently have researchers looked into shrinking the technology down to a microscale that can be threaded within the human body. The challenge has been to make the technology small enough to fit through human airways while using very small amounts of voltage to scan infrared light.

Xie’s prototype uses a MEMS-based (microelectromechanical system) approach, centered on a tiny, one-by-one-millimeter mirror. Xie and his students designed the mirror with tiny actuators, or mechanical supports, which pivot the mirror. As infrared light beams down the endoscope, the mirror steers the light back and forth, illuminating a slice of tissue. The reflected light bounces back up the endoscope, and is analyzed and depicted on a screen in real time.

The mirror can pivot 200 turns per second, at a 100-degree angle, enabling the scope to perform fast, real-time imaging. Xie tested the scope in rats, taking 3-D images of rat and mouse tongues.

The prototype is still too big to use in humans–it requires a total diameter of 5 millimeters to fit all its parts. However, Xie plans to further miniaturize the design, and will test the model in larger animals like pigs and goats in the next year. He recently started a company, WiOptix, and is seeking funding from the National Institutes of Health to help commercialize the technology.

Eric Seibel, research associate professor of mechanical engineering and director of the Human Photonics Lab at the University of Washington, says clinicians would have to be trained to interpret OCT images, which look more like ultrasound images than the visual images obtained from video cameras. He adds that size will determine whether OCT-based endoscopes work. “[This design] is a little bit more space-efficient, but it’s still more than five millimeters in size,” says Seibel. “It’s not quite there yet, but it’s a step in the right direction.”