X-ray mammography provides structural information, identifying breast tumors for further analysis. Gamma imaging, a functional technique, can help with diagnosing a lesion as a cancerous one. But if gamma cameras are going to be useful in clinical applications, they must overcome their low resolution (standard devices cannot detect tumors that are smaller than 1 cm) and unwieldy size. Using photomultiplier tubes from Hamamatsu Corp., researchers at the Thomas Jefferson National Accelerator Facility (Jefferson Lab) developed a compact gamma camera that is better suited to breast imaging applications. Designed for breast and small-animal imaging, the devices have fields of view ranging from 2 × 2 to 8 × 6 in. Intrinsic resolution has improved to 2 mm. The detectors also have a smaller dead space (the area between the detector housing and the imaging surface) than traditional gamma cameras. Although a space of about 50 mm is common, Jefferson Lab's devices have only about 7 mm of dead space, said Stan Majewski, head of the detector group in the facility's physics division. "We want to come as close as possible against the chest wall," he said, explaining how this enables easier breast imaging. Hamamatsu Corp.'s photomultiplier tubes give this gamma imager a 5 x 5-in. field of view and a dead space of 0.4 in. Courtesy of Dilon Technologies Inc. Hamamatsu's photomultipliers — small, compact, position-sensitive devices — have made the advances in resolution, size and dead space possible, Majewski said. The gamma imagers use an array of up to 8 × 6 photomultipliers. They have the high granularity needed to achieve good spatial resolution, along with intrinsic gain and a good signal-to-noise ratio. Although several other research groups and start-up companies also are using Hamamatsu photomultiplier tubes, some are basing their gamma imagers on the PIN or avalanche photodiodes. The signal-to-noise ratio and reliability of these devices is not as good, Majewski said. Perhaps more importantly, the cost of alternative photodetectors puts them out of reach for clinical instruments. Jefferson Lab — which is partnering with Dilon Technologies Inc., a start-up in Newport News, Va., to manufacture the gamma cameras — is guided to a large degree by economy. "Not everybody's driving a Volvo," Majewski said. "A lot of people are driving Chevys because it still gets you to your job." The R7600-00-C8, the Hamamatsu photomultiplier tube that the lab is using for most of its gamma imagers, offers good performance for the right price. Although the R7600-00-C12 is somewhat better for this application, Majewski said, it is about 20 percent more expensive. The real jewel, he said, is Hamamatsu's R5900-00-M64, which is considerably more expensive. But it enables imagers with a 0.7-mm resolution, so the researchers are likely to use it for imaging small-animal brains. Gamma detectors can provide functional brain images, but distinguishing various parts of a mouse brain — measuring 7 to 10 mm across — has proved difficult.