From an office in Tucson, Ariz., Dr. Ronald S. Weinstein, head of the pathology department at the University of Arizona College of Medicine, can see all the way to Lake Havasu City on the California border, a distance of several hundred miles. He not only can inspect things, but can do so on a microscopic scale.Weinstein doesn’t possess superhuman powers. Instead, he is taking advantage of telemedicine technology. Besides his other duties, he is director of the regional Arizona Telemedicine Program, which links 55 independent health care institutions, 71 communities and 171 sites. The program provides services in a variety of specialties, one of which is pathology, and runs its own data network, functioning much like a utility.For telepathology, the amount of information on an entire slide may be large, but all that must be transmitted at any one time is what is in the next field of view. The response, though, must not be slow if a pathologist wants to examine a nearby field of view. Although the program manages its own data transmission, it doesn’t supply all of the equipment or software. Part of that solution for telepathology comes from Apollo Telemedicine Inc. of Falls Church, Va., which offers work flow and application management software for telemedicine. According to CEO Mark J. Newburger, the company provides hardware if necessary, but focuses on the solution and software, partnering with hardware vendors, as appropriate. An “Orphan Annie” grouping of cells was imaged with telepathology. Named after the pupilless comic strip character, the term describes the way nuclei look within a cell, and the grouping often is related to thyroid carcinoma and psammoma bodies. Courtesy of Apollo Telemedicine Inc.The Arizona program has several of the Apollo Telemedicine systems, each of which operates in a real-time mode, with pathologists looking at frozen sections for tissue assessment much as they would in their ordinary practice. The systems also are used for second opinions -- in case a pathologist wants input from a colleague -- and for a pathology quality assurance program. Apollo Telemedicine has licensed patents for use in its products from Weinstein.In all of this, Newburger said that there is one overriding principle. “When you’re doing live interaction with a microscope, you want to approach as closely as possible the same experience that a pathologist has when they’re looking through the oculars of a microscope. In order to do that, you need to have a live image.”Part of the reason, he added, is the way pathologists typically work. They start by examining the sample at low power, so that as much of it is as visible as possible. After moving it around and looking at areas of interest, they use a higher magnification to make a diagnosis. Because it is impossible to know beforehand what will catch a pathologist’s eye, the only options are to image and store the entire slide at high power, use the virtual slide approach or make the experience of operating a system remotely as close as possible to that of operating a microscope locally. The latter is the approach followed by Apollo.Besides issues involving the network and the data rate, there also is a fundamental need to capture the image with great fidelity. Newburger acknowledged that camera technology does not yet offer the same resolution as would be obtained looking directly through oculars, but with the right camera, the technology can achieve more than adequate resolution.Apollo prefers a three-chip image sensor, which is found in an IK-TU51 camera from Toshiba Imaging Systems of Irvine, Calif. Image sensors are, by their nature, monochromatic, while pathology specimens are not. One could use a single chip, but getting color from such a camera involves placing a filter pattern atop individual pixels or using a filter wheel for the entire sensor. Both methods present problems. The filter pattern reduces resolution severalfold, and filter wheels cut data acquisition rates because three images must be captured instead of one.In contrast, using three chips avoids these problems. However, the three dedicated image sensors -- one for each primary color -- do increase the cost of cameras using this technology. And the pixels of the sensors must overlie and be optically aligned to one another, which requires careful optical engineering and camera construction.Gary R. Pitre, eastern regional sales manager for Toshiba America Information Systems, noted that the company has expertise in these areas. It uses a prism block assembly inside the camera and has accurate alignment of the three CCDs, he said.The camera used in the telepathology application has three 1/2-in. CCDs. The company’s three-chip cameras were originally developed for broadcast applications and are still used there, a fact that shows up in product plans. According to Pitre, Toshiba intends to introduce a high- definition version of its three-chip cameras sometime this summer, moving them from the current 768 × 494 pixels to a possible 960 × 1080 pixels.As for the future of telepathology, Weinstein would like to see two technical improvements, neither of which involves imaging technology directly. The first is the addition of an oil-immersion capability, something that the current systems lack but that pathologists employ when doing work locally. The second would give end users more choices and greater flexibility. “We look forward to the day when there are standards, and we look forward to the day when all telepathology systems are interoperable,” Weinstein said.Contact: Divya Mohan, Apollo Telemedicine Inc., Falls Church, Va.; e-mail: dmohan@apollotelemedicine.com. Gary R. Pitre, Toshiba America Information Systems, Imaging Systems Div.; e-mail: gary.pitre@tais.toshiba.com. Dr. Ronald S. Weinstein, University of Arizona, Health Sciences Center; e-mail: ronaldw@u.arizona.edu.