Do Not Fold; 14

Do not phone then staple or mutilate this card. The slogan of the computer a University of Illinois radio service presents a series of programs about you and the computer from banks to hospitals and from airlines to music. It's application in this team and these programs will give you a glimpse of these countless applications and what they mean to you. Do not fold takes its place in the modern hospital and on the physicians house calls Medicine finds an increased ability to aid mankind because of electronic data processing. Hospitals across the United States are now taking tentative steps toward computer based systems in all aspects of medicine and administration from medical records to dietetic requirements. The memories of computers spew forth information requested by men of medicine. Yeah.

The sound of an ambulance signals the approach of another patient at a local hospital on Friday observances must be ready for this incoming patient. His arrival triggers a pattern of action from admissions to the hospital laboratory at the moment. Many of these aspects of medicine are not relying on computers but the picture is changing. Dr. Ben Williams pathologist at Mercy Hospital in Urbana Illinois explains the most common computer system certainly the most widespread use of computers in hospitals as in many other areas including computer applications have been simply in the business office and of all of the hospitals in the country that have some access to computer systems. Most by far use their computers only for business type and accounting types of applications. A somewhat related application is one listed here and that's the scheduling of admissions.

This type of application is one that's great potential for the future because everyone is concerned more and more about rising costs of medical care. And one problem and in the rising cost of medical care is how efficiently can we use the facilities that we have. So that better means of scheduling patients for admission to the hospital for surgery or for the performance of various kinds of tests so that the personnel and facilities can be utilized more efficiently. It's potentially very important at the present time however there are certainly only a very few hospitals in fact using computers to schedule admissions. Another person who is deeply involved in the medical use of computers explains why such systems are not yet widespread. Robert S. Ledley president of the biomedical pattered foundation in Silver Spring Maryland comments. Well I think just that it's extremely important to do the obvious

things with computers in medicine a lot of these obvious things. The systems are already developed and in and in use and have been used for a number of years. In industry such as. Hotel reservations and line reservations credit card systems. One would think that it would be just as important to apply these going systems to. Medical records or medical information. The way it is now people travel a lot and if they get sick or something happens to them there's no medical record. No way of even getting by. They can always look in their wallet and find out their Or. Their. Status their financial status at the drop of a hat. It's a little bit ridiculous to a certain extent. Not to have someone. In the medical. Profession. One of the most promising areas in which computers may help the overburdened physician and hospital is

medical records from time immemorial doctors have maintained records of a patients visits and maladies. For years these records have been inadequate misfiled and unstructured facts have been tucked away in the mind of the position rather than those file drawers. Dr. Bennett Williams outlines the problem major difficulty with medical charts as they're now conceived an operational and in most hospital of a country is it. The information in medical charge has tended to grow as new techniques have arisen. And this information is generally very poorly integrated into any kind of coherent picture that can be rapidly assimilated by anyone who looks at the chart and doesn't already know what's in there. In other words. As a new technique of diagnostic study for example may have been developed

generally. The new page so labelled would have been I added to the hospital chart. And it's to be found somewhere in the chart but unrelated in time to other studies that were let's say the same day of the day before or after There's no there's no coherence in time to indicate at what point during hospitalisation particular studies were done or how the results of. These various studies relate to each other at one particular time. So that a. Major contribution could be made by computer handling of medical records. Just in the programming and the filing of data so that it could be called up. In a coherent way and preferably. With several options as to how it might be called up. For example at one time a position might be interested in the changes that occur in the results of a particular kind of test over a period of several days

at another time he might be more interested in the results of several different kinds of tests on the same day and it would be. Very useful to simply indicate one of these two choices through a terminal to the computer and have the display of the results come back and in the format that was indicated. This sorry state of affairs with manually prepared files not only hinders the doctor's treatment of the patient but the efforts of research on a high level. Without ready access to a vast number of files medical research becomes a cumbersome project. One of the fundamental current problems in medical records. Is that there are vast storehouses of medical records in hospitals all over the country virtually all of which is absolutely useless. In terms of retrieving the data from these records to analyze that data. To find patterns of disease because the record

is written and there is no way that a machine or any automated process can retrieve this data in the context that it was presented to the chart. Whereas if. The. Information was typed in be stored in a machine form one can for example I quickly retrieve all those patients who had pain and the right upper quadrant of the abdomen. And work females between 45 and 50 who had three or more children. This could be done and instant whereas it virtually impossible now. Furthermore it's this kind of data that is so essential to deriving our developing diagnostic programs. Where the machine itself can indicate the most likely diagnoses given the set of such symptoms the problem currently is that we just don't have the statistical data of the Association of such signs and symptoms that would permit. These kinds of

diagnostic logical conclusions to be written in a formal way necessary for the computer is done and now instinctively many times whatever instinctive means by the physician and this it becomes a large part of his training. But we don't have these. Written down in a formal logical way that can be used in a machine instruction. With easily operated terminals connected to a computer. Patients may quickly communicate their medical history and present problems. Questions posed by the computer may be answered by pressing appropriate keys on a typewriter. Thus the patient may be asked has anyone in your immediate family had heart problems. The patient may respond yes or no. Still later the patient may be asked to describe the presence of recurring pain and its location. If this pain is pinpointed to the proper location the computer program might immediately note for the doctor that the answers to these two questions suggest an investigation of heart problems during observation by the position perhaps in the hospital

changes and condition of the patient may be noted in this computer maintained file structured comments about quantitative descriptions of the patient and results of laboratory tests may also be placed in the memory of the computer laboratory techniques for making especially significant strides because of computer aids. The complexity of lab tests and the tremendous volume that must be handle each day challenges even the most efficient laboratory. Robert Ledley directs the National biomedical foundation dedicated to the development of new laboratory techniques aided by computer. He describes the analysis of photographs such as X-rays and enlarged photographs of blood cells. I think computer vision computer works. With numbers. Handel's number inside its memory. There are numbers. So our problem is to turn the picture. Into an array of numbers. And what we do is we take the picture and. Look at it as if

there was a graph paper don't want a picture on the graph paper as everybody knows breaks off a picture or would you break a surface into little boxes. Now each one of these little boxes has a certain darkness certain density and we give that darkness a number to something very white would be zero or something a little bit gray would be one and so forth until something very dark might be that same number seven. So we would have from 0 through 7 levels. And in the memory of the computer we put the American value of the prelim where the picture is recorded that you know Ray of numbers which represent the green levels in the picture. That's all right. This is all picture information. Why assign the analysis of these photographs to computers. Why not use the human eye. Such techniques seem so complex but Lyster Ledley easily justifies a computer

based system. The RIAA of course is very good in many respects. But the guy the people you know the AI is attached to a person and a person gets tired after a while and that particular applications we have in mind requires the analysis of many thousands of such pictures. For example when they send around the little truck that everybody can get their chest X-ray on somebody has to sit there and look at several hundred thousand X-rays which is very time consuming and very tedious and people make mistakes when they get tired and they don't do it right. So if we had a computer to do it the computer just won't get tired. The heartbeat of a man is an especially vital sign one that indicates his health already having like recent technological advances have made it possible to transmit the elements of a heartbeat over thousands of miles.

Experts may then study the electrocardiogram and determine the problem of the patient. This requires a. Significant amount of time to interpret each part tracings are like a crowded room waveforms. Which in the interpretation which has actually been quite successful in some large centers. Through a computer. And this is particularly significant when one considers the continuous monitoring of some patients. Heart tracings 24 hours a day when they are seriously ill or after surgery. In such a way a computer can analyze these data very rapidly virtually instantaneously as they are produced and by appropriate programming can itself make decisions as to when. There are conditions

that should call for an alarm. When i will there is a very early trend. Away from the. Normal course of events that one would expect so that a physician a nurse can be alerted much earlier than they might be if they had to wait until the clear signs of deterioration might occur. Even the kitchen staff of a local hospital benefits from computer systems. The elementary planning of menus for patients can be assigned to a computer. For example the several Mayo Hospital in New Orleans uses a computer to plan interesting and varied mails at a reduced cost. Typical diets food content and cost can be determined and less expensive but equally nutritious items may be substituted. In this one hundred and twelve bad hospital food

costs may be cut as much as 32 cents a day per patient without nutritional drop. This sophisticated computer program developed by the University of Tulane considers the patient's preferences dietary requirements nutrition cost. I appeal and variation of items. The computer program may also keep track of market fluctuations and seasonal items and detailed the amount of ingredients needed for each dish. No longer can patients predict Wednesday's menu or anticipate the next time scrambled eggs will be served at breakfast. In addition dieticians have an opportunity to develop new dishes because of time freed from previous menu planning by hand. Cause. Even the initial training of medical personnel receives a helping hand from modern processes of

electronic data processing. At the University of Illinois Medical Center in Chicago students are being trained in medical diagnosis by computer no longer to human guinea pigs suffer the mistake of the medical student. Mitchell Shahram former chief of the instructional systems section at the Medical Center campus outlines the laboratory and the students interaction with the program. Let me describe what it would be like if you were to come up to our laboratory you'd come up and you'd find four typewriters look very much like electric typewriters and you'd sit down at one of these and sign in with your name and be given a problem say a patient is perhaps forty nine years old has two years of low back home. And is not in acute distress. What you want to do. We tried to give three things to every problem what the problem is what the conditions are in which the facilities you have to work with and who you are. Because this is usually true and any practice. From

that point on you simply type in using the typewriter in your own language. That is using English misspelling punctuating are not. However you wish you know what you like to do if you want to talk to the patient to begin to talk to that patient talking and being used as a as typing. Well and then you just continue to manage the patient now as you get deeper and deeper into the problem you may start ordering laboratory tests or you may start some therapy. In any event you are given only the consequences of what your intervention yes and this will be typed out. Now there are some other gimmicks we throw in for example if you'd like to listen to the patients chest. We have stuff for phones. And you can and we have recordings. And so if you were at the patient's bedside using it using a stethoscope. In this case you can use the stuff a phone and we can retrieve that patient's

sound patient that. You're managing this interaction with a computer may seem stilted at first but the medical student readily adapts and accepts this dummy as a human being. There is an experiment that was conducted in an eastern school where they use natural language they had a person sit down had a typewriter and perhaps this is a scary part of it. And talk to a computer. I have a patient as a simulated patient as we say in the computer but they played a trick on the person they tried do they programmed by computer. The typewriter to type hesitatingly when it replied and to misspell occasionally and go back over and over type correcting it now this was built into the program as the computer executed it after the sequence was over they asked the person if they felt there was really a computer at the other end or if in fact another person was at the other and the few people they

tried this on they all said oh you can't fool us there's another person at the other end and not a computer. So it gets to be quite dramatic and it gets to be quite realistic. And here again we think is the is the payoff because this is very expensive and we can't justify it on a practical basis except if we get the kinds of answers we're looking for. And it doesn't please medical care then we can say can justify the costs. Dr. Williams of Mercy Hospital would extend this training even beyond medical school into the remainder of the Doctor's life. Perhaps the the most interesting speculate a way in which computer aided instruction might apply is in a very serious problem in medicine of continuing medical education and this is in many ways a far more serious problem than simply the education of doctors in medical schools all of. It as we've all been

hearing for a few years now there is. And imbalance between the number of physicians available to take care of people and the number of people who need to be taken care of in this imbalance is growing. But even apart from that is the problem of the very rapidly changing. Science of medicine and the problem of keeping physicians up to date as to what is new and the latest in the field certainly. This has been a problem that has received governmental attention over several years and it's not confined at all the medicine but the problem of the lag time between the discovery of new scientific conclusions are facts and the application of these conclusions. And facts to the actual practice of medicine. So one would hope

that computer based instruction. In a tie probably related to the system we use now. Plato could be could be effectively used to help physicians keep abreast of what is the latest in the field. Nurses are also being trained by computer. The PLATO system developed at the University of Illinois enables students to interact with special television and typewriter terminals to learn material in a particular subject. One of these subjects is nursing Mariette beds or for the staff of the Mersey School of Nursing explains the development of this project. Back in 1963 we did a pilot study using a computer controlled teaching system to provide a simulated laboratory for nursing students in the simulated laboratory they were able to do the experimenting and inquiring which allowed them sufficient flexibility to proceed in their own way in learning. Situations that would not be dangerous to the

patient. The student response of this type of learning was very favorable and we found sufficient indication that we should pursue this course of study. In September of 1966 Mercy Hospital was awarded funds from the Department of Health Education and Welfare to pursue this course of study. The maternity nursing course which was taught to two year students at Mercy Hospital. Was programmed and adapted for use on PLATO system. To date we've had about 80 students who completed anywhere from eight to twenty two of these lessons. We have also begun writing and adapting a course in pharmacology for nurses. Today about 20 students have completed these lessons. This is better also discusses basic concepts of the system. The studies which we have conducted to date show that the students have gained significant amount of knowledge in these courses and have completed the

courses within one third to one half the time required by those students in the conventional classroom. All of them have showed a significant gain in their post testing as compared to their pre testing. However there is not what we call a significant difference between the control an experimental group of students. The big difference lie in the time savings I believe. And as I said this is from one half to one third the time now with us one third to one half the time that you say the instructor has an opportunity to develop a different type of teaching. She can go into in-depth teaching. She can develop more of her clinical teaching. She can't have time that freed to do many of the things that she really wished she could do but has previously used up presenting informational content. So therefore we feel that there is a type of content in a course that has informational content that can be presented just as well by the computer as by an

instructor. And at the same time the student is allowed to obtain this informational content in a manner which is suited to her own level of ability she can pace herself. She can also determine the direction as well as the rate of her learning. Beyond the walls of the local hospitals computers and medicine work hand in hand to aid the average citizen. Community oriented projects that are underway provide medical aid to people and put the skills of medicine on call. For example many towns have set up poison control centers. Robert Lee explains how computers affect the centers. Most cities or every city really has a poison control center. Usually at a children's hospital. Now the parent who has a child who has taken some point and usually children take Portman's

calls off a doctor or the police or the fire department and they are there for the mother to the poison control center so she telephoned the Poison Control Center at the present time and there is an internal resident whence is the phone and he has a deck of about 2000 cards or 25 or cards and he tries to find the poison on this card. You know the cookie they have one card for each type of point tries to find the right poison. And if he doesn't find it on the cards then he looks in some books that he's got there. Usually takes about five or 10 minutes so he calls the mother back. And it's not very rapid. Not only that but the cards which are by the way given and kept current by the poison control clearing house which is run by the federal government. These cards are alphabetically organized so if someone is all sloppy and puts the cards in the wrong alphabetical order than mine is when I have a card. Now if you have a computer assist here. Let's say that the

physician at the Poison Control Center said it looking through a bunch of cards he had teletype machine and he just telephone the computer with a teletype and just typed only the brand name of the poison the computer could of course get a much bigger memory than 2000 cards they can look up the proffer he can find in its memory. This particular item and it can the send back a message to the physician. What the prevention or cure or treatment is treatment would be French in this case would be treatment for this for the poison. And then of course you can tell the mother right then and within seconds. Care Centers for a community may go so far as to check every citizen healthy or not. Early stages of disease may be discovered in routine tests such a volume of laboratory work is only possible with computer based systems according to Dr. Williams. One area was. That so considerable growing interest to

people concerned with the delivery of health care health screening. That is to periodically perform great variety again. With new economies of different kinds of tests on people who appear to be healthy. To detect disease in the very earliest stages so that it can be treated and possibly cured a fight is possible before the patient is even aware that a problem exists or if the cure is not possible at least to slow down the progress of the disease. Or to reduce the symptoms present and the disease. This type of approach. Is by its nature a concern with the testing of large numbers of. People on a periodic basis another related feature here is

that one can establish what is normal for any one individual one thing again in a much more sensitive way detect when. That individual is retested whether he has deviated more than he should have from the already established patterns that have been stored in the system. But this approach then involves by its nature the testing of large numbers of people. In a in complex instrumental way is to get large amounts of data that one hopes will have some medical significance much still remains to be done in the field of medicine before computers will become a part of an overall system. Isolated computers perform limited tasks but no hospital yet has a truly integrated system stretching from administrative to laboratory work and back to patient monitoring and medical history. This development is part of the future in the computer age.

Do not fold is the latest technique in medical research and promise for better care in tomorrow's hospitals. Computers may change the efficiency and capability of Medicine markedly in the next few decades. On our next program in this series computers match their wits with tornadoes thunderstorms and other aspects of the weather. Each week the University of Illinois radio service brings you a new meaning behind the slogan of the computer age. Do not phone have been a staple or mutilate this card. This program was distributed by the national educational radio network.