High Performance Computing and Communications

Donald A.B. Lindberg, M.D. Director, National Library of Medicine and Director, National Coordination Office for High Performance Computing and Communications

The multiagency federal High Performance Computing and Communications Program provides the technical foundation for the National Information Infrastructure (NII) or emerging "Information Superhighway." Health care is a key component of the government's vision for the NII, which will put the vast amount of resulting information at users' fingertips.

It is clear that the information infrastructure of health care depends on new computer and communications technologies. Advances in these technologies will have a wide ranging and beneficial impact on our nation's health, on the intellectual vigor of its people, and on its industrial productivity.

The National Library of Medicine (NLM) has a long and productive history of innovation in the use of digital communications networks. Thirty years ago, in 1964, NLM first created a computerized version of Index Medicus, an index catalog to the world's biomedical literature. In 1971, NLM established a contract with a fledgling company called Tymnet, to provide long distance digital telecommunications for online literature searching of that Index, via a new system called MEDLINE.

Today, MEDLINE and the more than 40 other databases that make up a system called MEDLARS (Medical Literature Analysis and Retrieval System), comprise the largest and most widely used biomedical computer system in the world; over five million computer searches were done on the NLM computer system this past year. About half of those searches were done for medical research and education, and the other half were to get information for direct care of a sick patient.

The NLM, under the aegis of its Board of Regents, undertook an ambitious long range planning effort in the mid-1980s. One outcome was recognition of the growing importance of electronic images as a source of biomedical knowledge; the planners' report specifically recommended that "NLM investigate the feasibility of building digital image libraries." In 1988 NLM convened a meeting of experts who observed that NLM might facilitate progress in this area by building a digital image library representing the x-y-z numerical coordinates of an entire human body, a project too large to be undertaken in any single university center. The Board of Regents approved the project and incorporated it into the NLM Long Range Plan in 1989.

Phase I of the "Visible Human Project" is now under way via a contract with the University of Colorado Health Sciences Center. Under the guidance of principal investigators, Dr. David G. Whitlock and Dr. Victor M. Spitzer, the project is now assembling image data from photographic, computed tomography, and magnetic resonance imaging of representative normal male and female cadavers. These data are expected to be available later in 1994 as a set of digital images that NLM will make available over the National Research and Education Network.

For practitioners who work in isolated rural areas, digital telecommunications may provide "agile medical care" via instantaneous video consultation with colleagues at distant major medical centers, or the simultaneous viewing of X-rays and discussion of findings while both practitioners see the images on viewing screens, perhaps thousands of miles apart.

In May, NLM announced the award of 12 contracts totalling $26 million designed to help physicians practice better medicine by utilizing advanced computing and networking capabilities. These include testbed networks to share information resources, computerized patient records, and medical images; telemedicine projects to provide consultation and medical care to patients in rural areas; and advanced computer simulations of human anatomy for training via "virtual surgery