Indiana University      Research & Creative Activity      September 1999 Volume XXII Number 2

Cohen

Has information technology enhanced radiology? Mervyn Cohen, Eugene C. Klatte Professor of Radiology and chair of the Department of Radiology at the Indiana University School of Medicine, thinks so. "In 1976, about the time I started in this field, computed tomography (CT) took two to three minutes to process each individual image. Now with high-computation abilities, image processing and generation can be done in virtually real time," Cohen says. The increased processing speed helps patient comfort and significantly improves productivity while decreasing costs.

Cohen divides radiology into three eras. The first era spans the time from the discovery of X rays by Roentgen in 1895 until the middle of the twentieth century. During this period, imaging of the human body improved significantly, but it did so without the aid of computers or information technology. The second era of radiology spans the latter half of the twentieth century, with increasing involvement of computers in human imaging with the invention and development of such imaging modalities as nuclear medicine, ultrasound, CT, and magnetic resonance images (MRI). In the third era, the past twenty years, significant advances have been made in the quality of images and the speed with which these images can be processed.

For all the advances that computing power has brought to radiology, Cohen is equally excited by two other aspects of computer and information technology: digitalization (conversion into an electronic format) of all human body images and the communication of those images over networks. "We now have the technology to convert any image including traditional X rays, ultrasound, and CT into a digital form," Cohen reports. "Within the next eighteen months we expect our radiology department to be completely filmless." Many methods convert body images into an electronic (digital) format. CT and MR images are directly acquired as digital information. Plain film radiographs such as chest X rays or bone X rays are now being acquired in a digital format using new technology that involves placing a special detector in the X-ray cassette instead of X-ray film. Pixels in this detector are excited by the energy of the X rays. This energy is released as a green light by a special processing system. A digital light sensor picks up the light rays from each pixel and generates a digital image. Although it works well, this process is relatively slow. A new technology now being added allows a direct digital detector to be placed behind the patient so that the image is instantaneously acquired in a digital format. The ability to acquire images digitally has multiple advantages. When X-ray film is occasionally over or under exposed, the patient must be X rayed again. Digital X-ray imaging allows manipulation of the contrast and brightness of an image, compensating for exposure problems. Other manipulation of electronically acquired images can significantly enhance the ability to detect disease.

Because digital X rays offer both faster patient service and improved diagnostic capability, it is no wonder the radiology department is moving out of film and into digital technology. Yet the greatest benefit, Cohen believes, will be in the network transmission of these images, available anywhere, anytime, and extremely rapidly. No longer do referring physicians have to spend time wandering back and forth to the radiology department. Doctors working in intensive care units have the images of their patients instantaneously available on viewing monitors in their work areas.

"Multi-site institutes, like ours, are becoming the norm," Cohen says. "IU/Riley and Methodist hospitals are now part of Clarian Health. That means there are several radiology labs, different clinics, and specialists at many sites around the city. What happens if two different doctors need the same image simultaneously?" With film, the X rays would have to shuttle back and forth. The solution is digitalization and networking. Any number of medical personnel can draw up the digital images of the radiological investigation at any time and at any place, as long as they have a computer with a network connection or modem. Cohen's department is working with the Regenstreif Institute and its medical records system to digitalize all radiology results and enter them into the larger electronic database.

Cohen foresees many advantages to such a system. Eventually, he thinks, the concept will be applied beyond local hospital networks to the Internet. "A general practitioner will then look at images on his personal computer and talk with a specialist, who has the image enhanced on a workstation screen, and thus through electronic consultation the technology will help improve diagnosis or treatment of a patient." In the future, the generalist and specialist can share the same image by accessing the Internet, no matter how far apart they may be physically and no matter how disparate their institutions might be.

Other advantages of networking electronic images are that specialists can quickly read all X rays, thus improving quality of care. The image report can be integrated with the electronic image so that the referring physician can speedily receive back both the patient's image and radiologist's report. Turnaround time can be significantly improved.

Cohen has a broad vision to apply information technology to other areas of everyday radiological practice. "One of the areas we are looking at for the future is work flow management. The radiology department typically has peaks and valleys in work flow, yet we must have personnel on duty all the time to handle the peaks when they occur." He expects that a computerized model, taking into account factors like bad weather, the scheduling of national medical conferences, and the frequency of radiological examination requests from each general physician, can predict high use and low use days and even predict high use hours or minutes within each day. Scheduling of staff and preparation of equipment can become more efficient with such a model in place. "We make a lot of these judgments by intuition already, but without completely empirical data. A computer model can regularize that knowledge and produce an accurate predictive schedule," he states. The radiology department is consulting with many IU departments including philosophy, computer science, engineering, and informatics to determine variables and programming for this type of work-flow model.

Cohen is working on other ideas as well. He believes computer programs can help with medical decision support, such as treatment guidelines and physician prompts. "We can develop programs that aid the examining physician and the radiologist," he says. His department collaborates with researchers in the Department of Public Health and in IU's School of Public and Environmental Affairs for long-range development of medical decision support programs.

To fulfill the vision of a comprehensive information system for all aspects, medical and administrative, of the Department of Radiology, Cohen is fostering the establishment of a new division of informatics. "We have just recruited a new director of informatics, someone from outside the field of radiology with extensive experience in informatics and industry. We are one of the first radiology departments in the world to have created such a full-time position," he says. "We believe that this will allow us to take the lead in information systems and technology in radiology, as we are looking to become an informatics leader." --William Rozycki

For more information on the Web:

Return to the Table of Contents