Introduction to the Paper by G. M. Kolodny, M.D., et al, “An All-digital Nuclear Medicine Department”

ALTHOUGH OUR ALL DIGITAL DEPARTMENT has gone through much change since this “classic” article was published in 1983, certain design principles have remained constant and have proved useful. These principles through the years have included use of standard off-the-shelf hardware and software, to not only lower costs but also to limit the time and aggravation of debugging software and service problems, and to make easy and seamless connections between many different systems. It was essential to us, and has continued to be essential to us, to be able to communicate with a wide range of nuclear medicine cameras from different manufacturers and to be able to display and analyze data from all of them. We were the first to use standard modems, then the Internet, and finally cable modems to do teleradiology in an ongoing routine clinical setting. This has been possible because of our insistence on using standard off-the-shelf hardware and software for connectivity, and not proprietary designed software and equipment. We have insisted on at least three levels of backup and redundancy—sometimes four. This has resulted in a miniPACS design, where each section of a radiology department has its own standalone PACS, that is able to communicate with any other miniPACS. This redundancy has proven itself as we have effortlessly stored over 17 years of data, with no meaningful downtime. Because of this design, all studies, no matter how old, are, and have always been, accessible in less than 1 second. A problem in one miniPACS has no effect on the operation of the rest of the department. As can be seen from the authors of our “classic” paper, and continuing until our latest 4th generation PACS, radiologists have been intimately involved in the design of every function of the PACS system so that “user friendly” is not just a marketing slogan. It has been essential that all workstation functions be self-evident, and that no training or instruction is necessary. We started with nuclear medicine because the data requirements for display, communication, and storage are relatively low compared to the other radiology modalities. As the general computer world and its peripherals have increased storage and display capabilities, we have followed with computed tomography, then magnetic resonance imaging, ultrasound, and finally plain film storage and display. This means that our software has always been able to operate with standard operating systems running on unmodified general-purpose computers. We have never had to turn to expensive proprietary workstations, but could load our software on any PC at multiple locations. Nuclear medicine was the first radiology modality to use computers for functional analysis, so it’s not surprising that nuclear medicine has led the way in terms of both picture archiving and communication system (PACS) and functional analysis. However, nuclear medicine does have requirements that put a greater burden on display functions. For this reason most commercial PACS do not have adequate nuclear medicine display functions, which include color, multiple independently variable cines, and functional analysis programs for cardiac, renal, gastric emptying, lung, etc. Also we have found the digital imaging and communications in medicine (DICOM) standard slow to adopt all the needed functions for nuclear medicine, which has necessitated using the Interfile format or being able to translate native formats of each version of the gamma cameras that have been manufactured. The future of PACS will see Java-based software that can run on any computer and that can be made available as shareware supported by commercial vendors, similar to the Linux model. We are now working with other investigators to provide our next generation PACS based on such a model. Nonetheless, the principles we developed for our first PACS design have stood the test of time across three versions of our PACS and will continue to guide us as we develop the next version.