| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ||||||||||||||||||||||||||||||||||||||||||||||||||
Special contributions |
Johns Hopkins School of Public Health, 615 N. Wolfe Street, Baltimore, Maryland 21205, USA
A recent meeting at the National Library of Medicine, sponsored by Johns Hopkins Medical Systems, the Society of Nuclear Medicine, the National Library of Medicine and the National Institutes of Health evidenced great enthusiasm for the opportunities to the field of nuclear medicine presented by the Internet. More than anything else, information technology can accelerate and further enhance the increasing fusion of nuclear medicine into the mainstream of medical imaging. Despite positron emission tomography (PET), nuclear medicine remains to a large degree a stepchild, off to the side of radiology, the main driving force of medical imaging, but the fusion of PET with CT and MRI is bringing about an integration of medical imaging. The Internet can be the vehicle enhancing this collaboration and cooperation.
Three institutions were clearly in the forefront of medical informatics in imaging: UCLA, Maryland VA Health Care System and Beth Israel Hospital in Boston. The UCLA group has been particularly effective in the integration of PET into radiology and hospital information systems.
For thousands of years, communication between a patient and those involved in his or her care has been the foundation of the practice of medicine. Our age has been called the Information Age, yet despite all the hype in publications and at meetings, information technology has hardly influenced the way most nuclear medicine physicians and radiologists practice medicine. Most of them still need to be convinced that modern information technology can help them care for their patients, increase their productivity and enhance their incomes. They do not realise that IT is a so-called "disruptive" technology, which will leave behind those who do not embrace it until it is too late.
Reliance on written transfer of information results in poorer care than is the case when physicians have personal contact with each other to discuss diagnostic tests, decisions and follow-up care with other specialist physicians and with their patients on a continuing basis. Studies such as PET are increasingly difficult to interpret and integrate with the information obtained with anatomical imaging, such as CT and PET.
When physicians are in the hospital, they often look at radiographic images, and to a lesser degree, pathology slides, together with the responsible radiologist or pathologist. This is usually not done in modern office practice because the physicians are too busy with their practice to take time to come to the hospital to view and discuss the images. By making transfer and simultaneous viewing of images routine, the Internet can enable high quality images to be displayed simultaneously for radiologists, nuclear medicine physicians, and referring physicians.
Eliot Siegel of the Maryland VA Health Care System described his system which in the year 2000 dealt with more than 60000 radiology examinations per year, an increase from 34000 studies per year in 1993, the year they installed their picture archiving and communication system (PACS). This increase in productivity occurred with an increase in the number of radiologists from 4.5 to 5.5, despite the 70% increase in the number of studies. The need to repeat radiographs rate fell from 5% to 0.8%, an 84% reduction. The time needed to return the reports to the referring physicians fell from 24 h to 2 h. In a survey of referring physicians, 98% indicated that the use of the PACS resulted in much more effective utilization of their time.
Under-utilized by the medical community today is a capability at the National Library of Medicine of a browser interface that allows nuclear medicine physicians to use their personal computers to view images at any remote site.
Gerald Kolodny described the Internet accessible imaging functions at the Beth Israel Deaconess Medical Center in Boston, based on the use of a cable modem (1.5 megabits/sec). This has been extremely effective in providing expert interpretation of images during the night and on weekends.
The unit of productivity of an imaging service is the number of studies per unit time. Ideally, the equipment and personnel should be in maximum use all times. While improving productivity, one must also insure optimum service to referring physicians, studies and interpretations of the highest quality, and an efficient flow of patients through the department. In face of the clear advantages of IT in delivering health care, why is its use not more widespread? Many physicians are afraid that information technology will result in having more guidelines thrust upon them by managed care organizations and hospital administrators. On the other hand, they believe that they have been unfairly tagged as technophobic because they have not jumped on the Internet bandwagon. To a large degree, the blame lies more with hospital administrators and information-system colleagues who focus all their institution's resources on financial matters with no thought as to how they might improve patient care. Furthermore, the top-down approach of administrators scares off physicians, who often view information technology as an enormously complex and expensive venture, beyond their reach until their institutional administrators take the lead and provide the funds to establish a system. For example, Warren Slack who set up the IT system at Boston's Beth Israel Hospital estimates that today to build their system from scratch would cost $50 million and take several years to integrate with other hospitals and systems. This top-down viewpoint makes physicians hesitant to get involved from the "bottom up". The message of the meeting at the Library of Medicine to the nuclear medicine community was "just do it".
Another inhibiting factor to the widespread adoption of PACS, introduced in the mid 1980s, was that they were vendor specific, and the vendors were reluctant to release information about their equipment. They wanted exclusivity and secrecy, which in some cases resulted in commercial casualties. Some radiology departments became unduly influenced by dominant vendors. Systems such as that resulting from the collaborative efforts of IBM and AT&T were shown at RSNA and then dropped in a few years. Fortunately, as a result of efforts of the Medical Diagnostic Imaging Support (MDIS) of the government, standardization was introduced by the development of DICOM (Digital Information and Computers in Medicine), first introduced at the RSNA in 1992. Having been burned in the past, when faced with today's new technology, most physicians still want someone else to go first. The Web changed everything.
We in nuclear medicine again have an opportunity and responsibility to lead the way to the use of the Internet in medical practice. We have been interpreting images on a computer screen since the late 1960s and have had filmless departments since that time. Today, Java makes it easy to interface a web browser with any computer, using a program that does not need to be installed by the user, giving us a built-in networking capability by using the Web to access the image server. We can now have easy integration of our own network with global networks. Java Security API allows security functions.
We need to show the benefits of connecting to others via the Internet, and show how information technology can improve patient care, productivity and increase their reimbursable services. We need to show them how they can set up such a system, how they connect to other persons and services, what it will cost, how it will save time, how it will improve their status, and how it can help deal with regulatory mandates. We need to move from being technology-driven to being clinically driven. We need to put the Internet to work in routine medical practice.
Medical imaging is no longer bounded by radiology. Whoever controls the patients will call the shots. We must consider ourselves specialists in clinical information management. The clock is ticking for nuclear medicine physicians who do mediocre work. "Dumbing down" nuclear medicine will not help patients. Integrating the Internet into clinical practice is no longer a luxury but a requirement for staying competitive.
Here are some key assumptions:
Many, if not most imaging procedures can benefit from consultations between the imaging specialist, referring physician and other professionals and medical subspecialists, often at dispersed sites. "A picture is worth a thousand words." For example, when images are viewed together, the degree of certainty (or uncertainty) in the interpretation becomes clearer. Decisions are optimized, and outcome is better.
Language is mankind's greatest invention; communication is mankind's greatest problem. From birth, human beings are vision-oriented. More than 90% of our sensory input is visual. Vision provides facts; language provides concepts. Both are needed for optimum actions.
The communication and image transfer system must be in regular use, and must enhance, not distract from the provision of critical medical care. An example is the care of a patient with chest pain in the emergency room. The system must be available 24/7 and not break down.
For hospitals without a nuclear medicine specialist, a technologist can perform the studies, with the images interpreted immediately by a nuclear medicine specialist or subspecialist at a distance. Images can be processed and analysed at the laptop just as at the computer console of the imaging system. Most state regulations permit infrequent consultations between licensed physicians residing in different states. Initially one may wish to provide only consultations and second opinions using tele-imaging. Consultation fees are acceptable and have been used by others
PET, SPECT, CT and MR images require storage and transmission matrices no greater than 512 x 512 picture elements (pixels). Our digital images can be transmitted without loss of fidelity over transmission lines that cost approximately $40 per month. For downloading these images, a modern laptop computer will suffice. PET, SPECT, CT and MRI images are digital during acquisition and data can be fed directly into an Internet server without the need for a digitizer, and be available immediately for interpretation. From a technological point of view, tele-imaging for nuclear medicine, CT and MRI is almost trivial. (Plain chest radiographs have high information density, and require high quality monitors with 2000–4000 line resolution. This consideration, and the fact that the X-rays must be digitized, has been an inhibitory factor in teleradiology.)
The main problem in implementing the Internet into the practice of medical imaging is satisfaction with the status quo and hesitancy to accept new things. Not everyone is comfortable with computers, and all of us must become trained to their routine use in daily practice. When first introduced into nuclear medicine in the late 1960s, our trainees would hesitate to enter the rooms where the first computers were located.
We need to demonstrate how the use of the Internet on a regular basis can improve patient care, increase productivity and save money. High-end technology is not needed. An adequate tele-imaging system can cost less than $10,000 with $750 allowed for personal training. We need to replace confusion and uncertainty with knowledge and convincing demonstrations, and data documenting improved patient care and cost savings through increased productivity. We need to provide hands-on training with the information technology equipment, including providing videotape to instruct further the users of the system.
Some hospitals will choose to have their own tele-imaging services, but many imaging specialists will prefer to develop their own systems employing certain services provided by Applications Service Providers (ASPs), such as image storage and distribution. This is a "bottom-up" rather than a "top-down" approach. If a hospital chooses a vendor with proprietary hardware and software, it becomes dependent on that specific vendor for even the most minimal modifications. Proprietary solutions tend to be easier at first, but are rigid and are particularly vulnerable if the vendor goes out of business or is taken over by a larger company. An alternative approach is to store patient images on the Internet or on a local intranet in standard format, with encryption for safety. Any physician with a computer, browser and viewing software (and the appropriate password and decryption key) can access and view studies from his or her patients. In the case of home workstations, storage is not a problem because studies can be retrieved from the server.
Radiology and nuclear medicine are not the only specialties that can utilize tele-images. Pathology and cardiology have similar opportunities, and are moving in the same direction. The bandwidth requirements (512 x 512 matrices) are the same for all of their images. Nuclear medicine and radiology imaging systems should be interfaced with tele-pathology and tele-genomics. A key question is whether these specialties will also adopt a DICOM standard.
Today, many professionals are involved in a patient's care. As he or she goes from office, to hospital, to their patient's bedside, and to their home, the primary care physician needs to be able to connect with those involved in their care. For successful practice, four abilities are needed: availability, capability, affability and marketability. The Internet can help increase all of these abilities, starting with the relationship with one's best referring physicians, who will be able to schedule their own patients' studies. One can picture the process: Every morning the medical imaging specialist calls up a list of the patients scheduled for imaging studies that day. A click of the mouse brings up their electronic medical records. A quick search of the literature provides answers to some questions he or she may pose in view of the patients' problems that will be addressed. Then a program is called up to see if there have been similar patients in the databank. When I was an intern, I was concerned that typing notes while "taking the patient's history might interfere with my relationship with the patient". I really didn't find this to be the case, but today the use of a "Palm Pilot" or similar device when speaking with a patient can eliminate this problem.
Early in my training, I learned the importance of going to the Radiology Department to consult about patient X-rays performed that day. Fortunately, persons of the ability of the late David Gould were always present and willing to educate as well as communicate. Today, as soon as an imaging study is completed, the processed images can be viewed by the imaging specialist and referring physician. In interpreting the studies, the imaging physician will have access to experts for primary or secondary interpretations. Such a system will make it possible to receive and analyse images from throughout the world.
Patients love to be connected 24/7 to their physicians via the Internet. They feel more involved in their care, and in this way the Internet facilitates and extends the physician/patient relationship. Examples of questions that a patient with a lung lesion asks are: Is it cancer? How long has it been present? Can I be cured by surgery? Can I make it through the operation? How long do I have to live?
Imaging specialists must have a firm and continuing foundation in clinical medicine. Their clinical orientation is what made it possible for cardiologists to take over nuclear cardiology and cardiac MRI in the future. Of course, continuing collaboration with experts in the fledgling field of medical informatics and computer/imaging sciences is essential. We need to remind ourselves that nuclear medicine has been filmless since the introduction of minicomputers in the late 1960s. Today, the Internet makes it possible to build medical practices of high quality and great size, based on the sharing of vast quantities of knowledge and information at low cost and great convenience. Within five years, nuclear imaging could be the leader in the field of medical information and communications. In an age of interactivity and networks, where we can communicate instantaneously anywhere, anytime, medical administrators will have to earn respect and loyalty in new, and as yet unfamiliar, ways.
A reasonable question is: why has nuclear medicine technology not diffused as much through the health care system as is warranted by its broad applicability? Like an earthquake, many forces are reshaping medicine that are happening out of sight, and will soon erupt on the surface. The Internet is just such a force. Often, when some important change has occurred, people don't say "I was for it" or "I was against it". They say: "I didn't know it was happening." Within 5 years, IT will be tightly incorporated into the practice of medicine. IT is not meant to automate, but to transform the practice of medicine. It will improve rather than interfere with the fulfilling personal relationships so essential for the practice of medicine. The low cost of the Internet is a major advantage. The end-users of the Internet in medicine are the patients and their physicians. Each must have more involvement in the creation of uses for the enormous capabilities of the systems to be most effective. End-users and system creators/managers must work together.
Fast, effective communication is essential in today's and especially tomorrow's practice of medicine. Instead of the interaction being primarily between the physician and the patient, there will be connected individuals, all focusing on the patients' problems with their specialized knowledge. It behoves all nuclear medicine physicians to become more and more connected with others. It is essential for the care of their patients. Their status in the future will depend not only on their expertise and personality but also on how well and how promptly they communicate with others. Fortunately, two of the major characteristics of nuclear medicine imaging are that the original data are digital from the start and that the bandwidth requirements for the transmission of the image data are far less than for conventional radiographs or pathology slides. One can avoid the expense of digitizing analogue images and the need for high bandwidth transmission facilities. Nuclear medicine can do for the Internet what it did for computers in the practice of medicine.
People can now be connected in ways never before possible, and in the next generation the Internet will be 1000 times faster than it is today. The practice of medicine will be changed as never before. Nuclear medicine can lead the way.
Summary
Footnotes
Lecture presented by Dr Henry N Wagner, Jr on being awarded Honorary Membership of the BIR on 20 June 2001. 
| ||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| BJR | DMFR | IMAGING | ALL BIR JOURNALS |
