On its own, a single-pixel camera captures pictures that are pretty dull: squares that are completely black, completely white, or some shade of gray in between. All it does, after all, is detect brightness.
Yet by connecting a single-pixel camera to a patterned light source, a team of physicists in China has made detailed x-ray images using a statistical technique called ghost imaging, first pioneered 20 years ago in infrared and visible light. Researchers in the field say future versions of this system could take clear x-ray photographs with cheap cameras—no need for lenses and multipixel detectors—and less cancer-causing radiation than conventional techniques.
"Our system is much smaller and cheaper, and it could even be portable if you needed to take it into the field," says Wu Ling-An, a physicist at the Chinese Academy of Sciences in Beijing whose work with her colleagues was published on 28 March in Optica.
The researchers' system still isn't ready to be used in medicine. But they have lowered the x-ray dose by about a million times compared with earlier attempts, says Daniele Pelliccia, who in 2015 made some of the first x-ray ghost images. A physicist at Instruments & Data Tools, an optics startup near Melbourne, Australia, he used a building-size source of intense x-rays called a synchrotron, but Wu's group made do with a compact tabletop source. And whereas early x-ray ghost images were simple pictures of slits cut into metal, the Chinese group produced outlines of a seashell and of initials etched into metal plates. They have made "images that look like images," Pelliccia says. "The potential payback, if it works for medical images, is big."
The key to ghost imaging is to illuminate an object with light that has passed through a filter with a known pattern, says Miles Padgett, a physicist at the University of Glasgow in the United Kingdom. On the other side of the object, the single-pixel camera takes a picture—nothing more than a gray square. To end up with an image, you do this thousands of times, swapping out the filter pattern for a different one after each exposure. Wu's group used a piece of sandpaper, which is partially transparent to x-rays, and rotated it to create a different pattern after each exposure.
A computer produces the final image. Because the computer knows the filter pattern for each exposure, it can calculate the image from variations in the sequence of gray pixels captured by the camera. The result, in theory, is an x-ray image as good as any today, but without a high-resolution camera or the intense x-rays needed for conventional imaging.
Researchers have already demonstrated simple ghost imaging systems for optical and infrared light, which rely on programmable filters, says Jeffrey Shapiro, a physicist at the Massachusetts Institute of Technology in Cambridge. A computer records and resets the filter pattern as the light source projects it onto the object and the single-pixel detector.
Using an infrared system, Padgett's group has shown it can ghost-image a methane gas leak. The group's industry collaborator, M Squared Lasers, based in Glasgow, is working to commercialize the system and hoping to sell detectors to the oil and gas industry as a cheaper way to detect leaky pipelines, Padgett says.
Making a computer-programmable filter for x-rays is a bigger challenge, Wu says, because x-rays simply stream through most materials. Because her group resorted to unprogrammable sandpaper, it had to use a high-resolution camera to measure the patterns. But you could imagine a commercial x-ray system in which the manufacturer prerecords all the sandpaper patterns, Padgett says. Then, only the manufacturer would need the high-resolution camera, and individual users could simply buy a single-pixel camera and use the sandpaper filters in a specified sequence.
In order for ghost imaging to be viable in medicine, Wu says, researchers must show that the total x-ray dose needed for an image is lower than with a conventional system. One ghost image requires thousands of exposures, and the x-rays add up. In addition, the more detailed the object—for example, a human body—the more exposures you need. However, Wu says the x-ray intensity per exposure can be made low enough that ghost imaging may come out ahead.
Doing so would be important, Shapiro says. "If you could reduce the amount of x-ray exposure that women suffer in getting mammograms, or in chest exams, that would be a big deal," he says. But image quality still needs to be improved, he says. "It's got to be a good image."