PITTSBURGH, March 30 -- Medical records have entered the realm of the electronic world through the University of Pittsburgh Medical Center's (UPMC) Image Engine Project, a multimedia electronic medical record system that combines clinical images with textual data stored in UPMC's Medical Archival Record System (MARS). It is a research effort designed to develop an integrated multimedia view of the electronic medical record.
Funded by the U.S. National Library of Medicine's High Performance Computing and Communications and National Telemedicine Initiatives programs, the project began in 1994 and is based at the University of Pittsburgh's Clinical Multimedia Laboratory (CML) in the Center for Biomedical Informatics. It is directed by Henry Lowe, M.D., associate professor of medicine, and director of CML.
According to Dr. Lowe, most, if not all, existing computer medical records systems store only text. But increasingly, medicine has become very image intensive, and clinical images have become an important part of a patient's medical records.
"This is a problem," Dr. Lowe said. "In most computer-based medical records systems, images simply can't be stored in the system -- instead 'hard copy' files that undermine the speed and ease of a computerized system are needed to store these vital pieces of information."
The Image Engine system, however, may ameliorate this problem. This type of computer system can store and retrieve visual images much the same way that traditional systems store text. Ideally, an image engine would not only be able to pull up a clinical image by patient name and type of image, but also by type of disease, stage of disease, and even specified visual elements present in the image.
The system allows clinicians to download a series of thumbnail images on the screen which are used to help the physician select which full-size images to retrieve for viewing. Currently, the system can acquire, compress, store, retrieve, display and manipulate many kinds of clinical images, including radiographs, CT scans, MRI scans, nuclear medicine studies, gastrointestinal endoscopy images, EKGs and microscopic pathology.
Through the collaboration of project technologists and medical records specialists at the University of Pittsburgh Cancer Institute (UPCI), a complete multimedia medical records system has been established for about 600 patients seen at UPCI.
"For now, Image Engine is being evaluated in the oncology setting," Dr. Lowe said. "We are looking to expand into other clinical areas in the future."
"We can retrieve clinical images easily and use them to inform patients about the progress of their treatments or show them responses of tumors to alternative tests they may wish to try," said Ted Logan, M.D., assistant professor of medicine, medical oncologist at UPCI and collaborator on the Image Engine Project. "All of this is done while maintaining patient confidentiality."
"We can reuse images in the clinical system for teaching," Dr. Lowe explained. Designing the system to retrieve medical images without patient identification information helps expose medical students to the subtleties of images as they are obtained in real practice, while maintaining strict patient confidentiality.
Additionally, a researcher could scan a hospital's records for patients with a medical condition being studied and correlate that condition with any number of diagnostic or treatment results.
Dr. Lowe explained that while the system is very secure, as it requires logons and passwords at every session and is firewalled and isolated from the Internet, ensuring patient confidentiality in a variety of situations is a challenge. "Another part of the project is aimed at making data available in a very secure fashion," he explained. "We're designing the system to provide very secure access to authorized users from anywhere in the world."
In effect, clinicians would be able to consult through use of the Image Engine by simply taking a particular image and sending it through e-mail. This could be useful, according to Dr. Lowe, in situations where patients are seen by physicians in emergency situations while traveling out of state. Internet access could also facilitate collaborative research projects or telemedicine initiatives.
Recently, the Image Engine Project was awarded semi-finalist status in the Global Information Infrastructure (GII) Awards Program. The GII Awards recognized champions from all industries and sectors of society who use the Internet and network technology to produce amazing results.
"The Image Engine Project's achievement demonstrate that the Internet can lead to a society that is healthier, better educated and more prosperous," said James Hake, GII Awards chairman and founder. "As a GII semi-finalist, the Image Engine Project has been distinguished for innovation, producing real and valuable results and powerfully demonstrating the potential of the Internet."
For additional information about the Image Engine Project, please access http://www.cml.upmc.edu. For additional information about UPMC Health System, please access http://www.upmc.edu.
lk\3-30-98
###