 |
HPCC technology provides a potentially huge payoff in health care. From teleconsultations to availability of health related databases, the effect of using high performance computing and communications will enhance patient care, improve drug design, and broaden access to medical information. |
Current work on Chagas disease -- a parasitic illness -- is an example of successful health care collaborations. Biochemists, computer scientists, and computational chemists combine rational drug design methods with virtual reality and video teleconferencing technology to evaluate potential drugs for Chagas disease and related parasitic illnesses. In another effort, a magnetic resonance imaging resource center recently made its instrument available through the Internet, supporting the concept of laboratories without walls (collaboratories). Real-time imaging is also a growing capability in many medical centers, allowing physicians to use scientific visualization techniques to address health care problems.
Distant clinical teleconsultation is possible through a project that uses wide area communications supported by an infrastructure that incorporates local and remote storage. Standard interfaces allow multiple systems to collaborate in image acquisition and storage.
The health care industry is actively engaged in using HPCC technology. The Medical Connections program provides "jump start" funding to academic medical centers, community hospitals, and other health care organizations to connect to the Internet. Over 170 U.S. medical schools and health care facilities have been connected during the past three years. A program is available through three year grants designed to help physicians practice better medicine through use of advanced computing and networking capabilities (telemedicine). Analytic studies and the development of computer-based systems are also studied to promote the widespread transmission and use of medical records and clinical images in patient care.
A current telemedicine activity involves developing the Imaging Science and Information System (ISIS), which has the potential to detect and classify tiny precancerous formations (microcalcifications) in the breast, and making this facility available to area and regional patients via telecommunications. Other work uses the Multisurface Method (MSM) based on linear programming and multiple separating hyperplanes to analyze fine needle aspiration for diagnosing breast cancer. MSM may achieve 97% accuracy without the need for an experienced oncologist.
In a computer-based project, the Medical Literature Analysis and Retrieval System (MEDLARS) was established to achieve rapid access to the National Library of Medicine's vast store of biomedical and health-related information. MEDLARS search services are available on-line to individuals and institutions through world-wide communications networks, from which it is searched more than 18,000 times a day. Another project supports the development of: (1) three-dimensional interfaces to medical knowledge, including several large databases; (2) prototype vehicles for making current medical knowledge available to the general public via community networks such as freenets; and (3) prototypes for smart software agents to interact between medical knowledge servers and clinical information systems or computerized patient record systems.
The Visible Human data sets, consisting of images from a male cadaver and a female cadaver, have been completed. The data are detailed atlases of human anatomy at unprecedented resolution, created from thousands of images of a human body collected with state-of-the-art radiographic and photographic techniques. The larger, long-term goal is to produce a system of knowledge structures to form symbolic knowledge formats. This activity creates a virtual environment for those studying and researching human anatomy.
 |
On the workstation, this prototype visual medical chart integrates multidisciplinary medical images, including diagnostic-quality digital radiographs, with scanned documents and the patient's clinical data. |
 |
 |