Founder of the HELP System and the Utah Medical Informatics Program 2005 Interview of Homer R. Warner, Sr. Interviewer: Dean F. Sittig Editors: Rebecca M. Goodwin, Joan S. Ash & Dean F. Sittig Founder of the HELP System and the Utah Medical Informatics Program 2005 Interview of Homer R. Warner, Sr. Conversations with Medical Informatics Pioneers: An Oral History Collection Interviewer: Dean F. Sittig Editors: Rebecca M. Goodwin, Joan S. Ash & Dean F. Sittig Published by the U.S. National Library of Medicine, April 16, 2015; Updated April 8, 2015; updated , 2015. Copyright Information Government information at NLM Web sites is in the public domain. Public domain information may be freely distributed and copied, but it is requested that in any subsequent use the National Library of Medicine (NLM) be given appropriate acknowledgement. Suggested citation: Warner HR. "Founder of the HELP System and the Utah Medical Informatics Program," Conversations with Medical Informatics Pioneers: An oral history collection. Sittig DF, interviewers. Goodwin RM, Ash JS, Sittig DF, editors. Bethesda: U.S. National Library of Medicine (NLM); 2015. Available from: http://lhncbc.nlm.nih.gov/project/ medical-informatics-pioneers Photographs: When using NLM Web sites, you may encounter documents, illustrations, photographs, or other information resources contributed or licensed by private individuals, companies, or organizations that may be protected by U.S. and foreign copyright laws. Transmission or reproduction of protected items beyond that allowed by fair use Content distributed via the NLM Conversations with Medical Informatics Pioneers Oral History Project may be subject to additional license and use restrictions applied by the depositor. Oral History Interview with Homer Warner, Sr. Editors: Rebecca M. Goodwin, Joan S. Ash, and Dean F. Sittig Interviewers: Joan S. Ash and Dean F. Sittig Managing Editor: Rebecca M. Goodwin Copy Editing: Cindy Allen and Marie Gallagher Design & Layout: Troy HIll, Bona Kim, John Harrington and Anne Altemus Photo Editing: Troy Hill and Sandy Triolo Photo & Captions: Homer Warner, Jr., Melissa Rethlefsen, Heidi Greenberg, Mike Thelin, Stan Huff, Stan Clark, Reed Gardner, Renee E. Ziemer, Dean F. Sittig Deeds of Gift: Rebecca Warlow, John Rees and Jeff Reznick Contact: wwwnlm@nlm.nih.gov iv Conversations with Medical Informatics Pioneers: An Oral History Collection Homer Warner in the computer room at Intermountain Health. HELP was expanded throughout much of LDS Hospital over the byproduct, and a robust database was developed. Founder of Homer R. Warner was born in 1922 at the Latter Day Saints (LDS) Hospital in Salt Lake City, Utah and the HELP diagnostic laboratory in 1954. In 1964, Dr. Warner System and the Biophysics and Bioengineering department in the Utah Medical College of Engineering, which was renamed and relocated to the School of Medicine in 1972. In 1985, Informatics in the United States.1 In 2006, the department was Program renamed Biomedical Informatics. Dr. Warner and his Homer R. Warner, MD, PhD, was a cardiologist 2-7 He wrote who pioneered many aspects of using computers to augment the practice of medicine, including clinical decision support tools of medical informatics.8,9 of Utah and a doctorate degree in physiology from INTERVIEWED BY DEAN SITTIG MAY 16, 2005 second recipient of the Morris F. Collen Award of SALT LAKE CITY, UTAH May 16, 2005 1 Excellence.10,11 The us a little bit about where Homer R. Warner you were born and where award, named in his you grew up and went to honor, is presented school, and then how you each year at the AMIA became interested in this the paper that best HW Well, we're fairly describes approaches to close to where I was born. DS knowledge data HW Yeah. I was born in Salt Lake at the LDS management, and Hospital, as a matter of Homer R. Warner, a pilot in the VF "Grey" Squadron 199, in fact. On my mother's side, I came from a of these approaches. family of doctors. In fact, my grandfather was one of the founders of the Salt persisted, and the Department remains a leader in Lake Clinic. I think Mother always pictured me as informatics research, training, and implementation.12,13 coming home and setting up practice in the Salt Lake Clinic. I guess she was a little medicine, on the patent called On my mother's disappointed when I went into research. That was sort side, I came from a of the plan, you know. I went 1, 2008, the references list family of doctors. all through my schooling here in Salt Lake City, including medical school. Before I went he collaborated.14 Dr. Warner to medical school, I was in the Navy Air Corps for a illness.15–19 DS DS It is Monday, May 16, 2005, and this is Dean Sittig. I'm here with HW It was a Grumman Homer R. Warner, Sr., F4F-4, and I was a Navy in Salt Lake City, at his house. I guess I DS Oh, my goodness. have to say "Senior," Was that in the Second because your son is a lot of the people HW Yes. So I came working now probably home and went to know your son better medical school, and our than you. after the war, and we HW Probably. had a bunch of old men DS So I'll put the tape in there. We got through recorder up here, and in three years or so, on then we'll start asking an accelerated schedule. some questions. What I became interested in I'd like to start with is physiology, particularly. this: Could you just tell I had a particularly Earl H. Wood, 1950. 2 wonderful professor, Horace I did my residency the pressure wave from the aorta. W. Davenport, who inspired Dr. Wood liked it and thought I me to take an interest in that in internal should make a PhD dissertation area. After medical school, I medicine and then out of it, so I went back up to Minnesota for another year and went to Dallas and interned moved, for part of under Tinsley [Timothy] R. Harrison, who later that training, down requirements and so on, and got my PhD. So, here I was with an became Editor-in-Chief of to the Mayo Clinic. MD and a PhD and no job, and Harrison's Principles of I knew I wanted to come back Internal Medicine-for the They were doing to Salt Lake City. I was able to apply for an American Heart going to leave there that year Fellowship, and I got that, a one- and go to Alabama, so he bypass surgery. year appointment. sent me up to the University I came back here to the of Minnesota, where I did University of Utah, which my residency in internal medicine and then moved, was housed, for the clinical years, in the old County for part of that training, down to the Mayo Clinic. It Hospital [at 2100 South State Street]. I worked under was down there that I got interested in research. That a man named Hans Hecht. Dr. Hecht was a very bright would have been in about 1951, man, very enthusiastic. I right at the height of the initial I spent one night enjoyed that very much. He analyzing a single entrance into cardiovascular was hard to understand-he surgery, you know. They were had a thick German accent- heartbeat into but he was very good. During surgery. Earl H. Wood, whom I that year, I had a visit from worked with, was doing a lot of its harmonic the Administrator of the LDS the original work on diagnostic methods, with indicator-dilution components. Hospital, and Dr. W. Ray Rumel, who was the only curves and so on, to diagnose heart surgeon in town, asked congenital heart disease. It was just a very stimulating me if I would come to LDS and set up a diagnostic environment. I knew because I attended research cardiovascular laboratory so they could do heart seminars during the surgery. And so I did. summer that year, The next year, we went while I was on one of up there and built this the services. I went to laboratory patterned these seminars where after the approach that they reported every Dr. Wood had had at week on the progress of the Mayo Clinic. We their research. I decided did diagnostic studies, then that I wanted to primarily on congenital give research a try, and heart disease in those so I took a year out to days, because there do research, and never was a big backlog of got back! [laughter] patients. There'd been no So my thesis was on treatment for them, you the development of a see, until just then. So it method for estimating was an exciting time. the stroke volume of the But I had an interest heart on a beat-by-beat in research, and knew basis from the shape of this was what I really Homer R. Warner, 1951. May 16, 2005 3 wanted to do. I took a class while I was doing this peripheral pressure waveform to be more "peaked." work with diagnostic cardiology, at the University, in He told us that those kinds of systems could be what they called advanced engineering math. It really represented by electrical circuits. So I decided to build wasn't advanced at all. For me, it was very advanced, an electrical circuit that would be an analog of this but it was differential arterial bed. That got us equations. The teacher started on what turned was a man named C. out to be the analog R. Wiley, who wrote computer. We found out the textbook Advanced that we could indeed, Engineering Math in by using a second-order 1951. He was excellent. differential equation, He just had all kinds of build a system that could interesting illustrations distort the pressure of the principles he waveform exactly the was teaching. One way the arterial bed day, he talked about distorted the pressure Fourier analysis. I was waveform. very interested in that, I remember we put because of dealing a patient on the table with arterial pressure for diagnostic studies, waveforms. So I went and had a catheter in home, and I bought a the aorta. Then we put three-foot-long slide a needle in the radial rule with trigonometric artery for other reasons, functions on it. I spent Homer R. Warner, ca. 1960s. and while my helper one night analyzing a watched the patient, I'd single heartbeat into its run upstairs, and we'd harmonic components, and convinced myself that I feed the upstream pressure [aortic] waveform into understood how to do that. A few days later, he talked the circuit. Then we'd look at the output of the circuit about another technique, which he called a "transfer on one beam of a dual-beam oscilloscope, and on the function," which was able to use signals representing other beam, the output at the other end of the artery both the input and the output [radial artery], and then tune signals from a system, and said So I decided to the circuit until those two that you could mathematically characterize such a system. So build an electrical waveforms matched. At that point, then, we could read off I thought, "I'll try that." I'll circuit that would the parameters that described be an analog of do it on the same heartbeat on the physical properties of that the upstream and downstream artery. Well, because of that pressure wave, because during this arterial bed. research work, we got us our my thesis work, I knew that the 20,21 pressure wave got distorted as That got us started DS I'll bet! it traveled down the artery, but I didn't know how to explain it on what turned out HW And we used that analog or take care of it. I was always to be the analog computer, I expect, for about interested in it. So we did that, and we found that the system- computer. 10 years. That would have been back about 1960. We did the arterial bed-was acting all kinds of studies, and our like a resonant system; it had certain frequencies that interest, of course, was primarily with the circulation, because that's where we had access to real data. So being damped out. The resonant feature caused the we were interested in how the cardiac output got a transient drop in blood pressure, and a big drop in regulated when you exercised. We knew that there was a baroreceptor in the carotid sinus, and so we began cardiac output. So, our postulate was that the drop in studies to try to see what the transfer function of that resistance was responsible for the increase in cardiac system was, where the input would be pressure and output. If that were the case, if we could hold the the output was the frequency resistance constant by using The notion Burgess N. Christensen was that peripheral on the descending aorta and keeping the resistance at my student doing that project. resistance the resting level, we could controlled the He later became Professor of prove that regulation of Physiology at the University cardiac output during exercise of Texas in Galveston. But we cardiac output; it was accomplished through found that we could, indeed, regulation of pressure in the describe that transfer function, was perhaps the arteries. Indeed, we should be a quantitative description most interesting able to prevent the rise in heart rate and cardiac output upon of the transfer function. We thing I ever did. initiation of exercise. In fact, found it was sensitive to the that turned out to be the case. derivative, or the rate of change So, we published those results.22 of pressure, as well as the pressure and that it was a If I had to point to one thing that represents my nonlinear relationship, and, in addition, the transfer contribution, I suspect that's the notion that peripheral function "adapted" over time. Thus, the carotid sinus resistance controlled the cardiac output; it was perhaps was a device not for long-term regulation of blood the most interesting thing I ever did. pressure, but for quick adjustments, like during the onset of exercise, or DS Yeah, I'll say! going from a sitting HW It was an exciting to a standing position. time. In the meantime, So, that was the the digital computer came beginning of it, and along. The University then we studied other of Utah had acquired a components of that digital computer in the system. Engineering School, and We actually used I had a student working the analog computer with me-an intern, as the tool to control actually. We would go up experimental animals. to the computer and try running different kinds on the ascending aorta of programs. I wanted to to measure the cardiac learn how to use it. So we output. We'd place a Homer Warner with the analog computer. did, and we did a lot of catheter in the aorta to studies trying to describe measure the pressure the distribution of transfer recording in a dog, and times around different components of the circulation then we'd place a cloth cuff around the descending with dye indicators. We found some very interesting aorta just above the diaphragm. Between the cuff things, both in terms of the effects of different kinds of and the artery was a balloon connected to a tube activity and different kinds of drugs, and the shunting coming out of the chest of the dog to a pneumatic of blood from one organ system to another.23 happened was, we started the treadmill. There was May 16, 2005 5 And then one day, Robert Stephenson, a friend and We'd use that LDS Hospital laboratory. It was really, I think, largely as a faculty member in the data, then, to result of the diagnostic study Engineering School, pointed out an article by Robert S. create a matrix and some of the indicator- dilution studies we'd done Ledley* and Lee B. Lusted of symptoms and with the computer that made diseases and about using probabilities us eligible for a Research and logic to make medical Resource grant from NIH. diagnosis.24 They had used And that's how we got our Bayes' Theorem to assist with medical diagnosis.25 We Hospital. That was an exciting thought, well, that might be a good thing to try, because and so on. Once thing.28,29 We got a little IBM 1620 we've got a lot of patients we created that, computer to begin with, and coming through our laboratory with congenital heart disease. then we began then within a couple of years, we upgraded to a Control So we began collecting data to study new Data Corporation [CDC] patients to see if by asking each of the doctors 3200 computer. The 3200 who referred us cases for was a wonderful machine we could actually for its time, but there was no a checklist of the patient's software for it. It didn't even manifestations, and also to put predict a diagnosis have a FORTRAN compiler. down their own estimate of what they thought the patient on a given We did everything in assembly language. We allowed each had. And so we'd then send the patient before user only 2K of memory for patient through our diagnostic procedures. After the diagnostic the diagnostic their program. So the user had to write the program with its procedure you'd know what the procedure was own overlay, so it would bring done. real diagnosis was. It was a nice in more memory from disk. thing to do, because we had But with that, we accomplished data. We'd use that data, then, some interesting things. We to create a matrix of had students like Al Pryor, symptoms and diseases for instance, who did his PhD thesis developing computer-based ECG on. Once we created interpretation. We had that, then we began to Reed Gardner,* who study new patients to developed methods for see if we could actually monitoring patients. predict a diagnosis on a He implemented the given patient before the algorithm that I'd diagnostic procedure was developed in my thesis done.26 We published for estimating stroke that in the Journal of volume as the basis for In April 1965, Reed Gardner and Homer Warner presented the American Medical Association27 - that function at Fall Physiology Meetings in Atlantic City, NJ. In operating room, and later 2015, Reed recalls, "This photo was taken at a 25 cent photo was in 1961. But that booth! You should note that when Dean Sittig conducted this in the ICU. Those were was a breakthrough interview, Homer was 80 and what he remembered is right very exciting times.30,31 for us, because it drew It was a wonderful from the librarians. That is remarkable to me, especially now some attention to the that I am approaching that age." time, because you really 6 didn't have to read the literature. Nobody could have yellow light indicated there was some kind of a trend accomplished what we did then, because there was in one of the variables, maybe heart rate, cardiac no tool available. It's hard to imagine these days, that output, or peripheral resistance, for example. The we could have been in that situation. I stayed awake nurse then could press that light, which interrupted the a lot of nights trying to think of what to do with that computer and displayed a graph of what the trend was. machine, because we had all that computing power, The red light, of course, was an emergency, a cardiac and a few very capable people arrest or something, requiring We did everything to work with it, but it was a immediate attention. real challenge. One day, I was down in the Reed [Gardner] built in assembly ICU, a 10-bed post-operative computer terminals for us that open-heart surgery ICU, and used a memory oscilloscope, language. there was a nurse at the bedside because we couldn't afford the taking the blood pressure with machine cycles to regenerate the images on the screen. the cuff on one arm while in the other arm, we had Then, he had a bank of lights in the ICU-three lights a catheter recording the arterial blood pressure. I for each patient-a red, a yellow, and a green light. watched her, and she came back over to the nursing The green light would tell the nurse that the computer station, and I said, "Why were you taking that blood was sampling the patient's physiological data, so don't pressure with a cuff and stethoscope? We've already disturb the patient until that green light goes out. The got it recorded right here." She really didn't know the Dr. Warner with his family. Front row: Homer's wife Kay Warner, Homer R. Warner, son Willard "Wid" on his mom's lap, daughter Jodi on Homer's lap. Children standing in rear from left to right Katherine "Kathy", Stephen "Steve", and Homer Warner Jr. May 16, 2005 7 answer. She was frustrated. What it really amounted so many milliseconds to each user and switching to is, we'd overwhelmed her with information. We automatically amongst the users. But we couldn't do had all this data, but we didn't know what it meant. that-we were sampling data sometimes at 200 times So anyway, with this particular patient, there were per second with an electrocardiogram, for instance. some hemodynamic disturbances, and we called the If you missed the timing on that, you distorted the resident, who came down. The signal, so there's no way you patient had surgery a couple of I stayed awake a could tolerate that kind of a days before, and we thought the patient probably had a lot of nights trying system. This lovely computer that we had, had an interrupt cardiac tamponade. We called to think of what system that Al Pryor was able the surgeon, and he agreed, so to incorporate into our time- they took the patient back and to do with that sharing system and make it so took care of it. And I thought, wouldn't it be nice if we could machine, because that there were priorities to the different users based on how store that thought process we'd we had all that urgent the data was. been through so the next time that set of events occurred, computing power, So anyway, lots of people contributed to the building of we'd recognize what it was, and a few very that HELP system. It became we could help that nurse, and she wouldn't just have to try capable people a showpiece for us. We had visitors from all over the to process the data without any to work with it, world come to see what was help.29 happening, how we were doing That was the beginning but it was a real it, and so on. It became a tool of the HELP system. With that idea, we went ahead and challenge. for graduate students like you to work on.2,5,6 the HELP system. I wrote the code for that system DS Right, exactly. while I was in Europe. I had two meetings in Europe, HW Lots of bright young people could use that kind and I had time between those two meetings in Geneva. of an infrastructure to do their research. It opened up I had my mother with me, I took her on this trip, and the whole clinical domain for us and gave us some she'd never been to Europe. So here was Mother in credibility with the clinicians, the hospital, and the the hotel room with me. She was a very quiet little hospital administrators because lady and didn't bother me a bit. LDS Hospital had become a But I worked night and day, It was a system "pioneer" in the application of and I wrote the code for that computers to clinical medicine. system and brought it home, They were all tickled with what and we implemented it and time, incorporated we'd done. So that was a very published it. But that was the beginning. Of course it wasn't decision logic exciting period, I think. DS I'll say. anything like the real HELP system in use now, but it was into the database, HW And then it's still going so you had a on. They're now working incorporated decision logic into the database, so you had knowledge base as on new versions of HELP, new hardware, and they've a knowledge base as well as a well as a database. incorporated all the outpatient database.32,33 needs and so on. Order entry Of course, Al Pryor came has always been a tough thing. We didn't attack that along and built a very unique kind of time-sharing problem initially, and I think that's probably one of the system that made it all possible. Up till that time, the wisest things we did, to recognize early in the game time sharing was largely one of just simply allotting 8 that physicians aren't going to be very happy about turned out that in getting his research done at Tucson, entering data into the computer. But they're happy to he used much of the same technology we were using. get the results and analysis of those results back. The whole secret to it the sun, using video methods Wouldn't it be was having the right people. linked to a computer; we were We've had a crew; you look at using the same kind of methods people like John D. Morgan, nice if we could to follow variations in the left for instance. Here's John, ventricle after injection of a who comes in as a graduate store that thought radiopaque, to follow the dye student and did his thesis on automating the laboratory and process we'd been motion and characterize normal and abnormal contractions in helping to automate some of through so the the coding in medical records. Then he stayed on as faculty next time that set in quite well, and within a very short time he was speaking for a few years, but he wasn't of events occurred, "medical-ese" and talking to we'd recognize really an academician. That doctors, and he made friends isn't what he wanted, so very easily and made very nice he quit and started his own what it was, we contributions in our department business, invested his own before he left to go to personal money into a little could help that Columbia. It's fun to reminisce minicomputer and went around the country showing this to nurse, and she about people, the bright young people. Pretty soon, most of the wouldn't just have Informatics. hospitals in the country were using his system. He turned to try to process DS Yeah, you've had a lot around and gave us a $100,000 the data without of really good people come through here, that's for sure. any help. grant, which we're still using interest from to help support an HW Yeah, lots of good people. outstanding graduate student DS So you've answered a lot each year. So we've had some of the questions I was going to talk about. wonderful success stories, bright people who've come along. HW Well, I have probably just said everything I Now Paul Clayton's left us. [Reed Gardner notes: know! He originally left to establish the medical informatics department at Columbia, then came back to DS No, you haven't. You've done a few more things Intermountain, then left again to go on a mission.] The you might have forgotten about. So tell me about when you created the Department of-I guess story of how he came to me might be sort of interesting. The whole secret Biophysics and Computing. DS Absolutely, I'd love that. It sounds like you were really to it was having doing physiology or what I'd HW One day, he came to my lab; somebody had told him the right people. call bioengineering research before that. what we were doing up there. HW Oh, yeah. Originally, we astrophysics at the University of Arizona. He didn't called it Biophysics and Bioengineering. And that have a job opportunity, so somebody said, "Well, you came about when the Dean of the graduate school need to go talk to Dr. Warner up at LDS Hospital." was Henry Eyring, a world famous chemist. He was So he came up, and we had a chat, and he told me a good friend. He'd asked me on several occasions to what his thesis was about, how he'd approached it. It make presentations to his graduate students on some of the things we were doing. One day, we rode home May 16, 2005 9 from a trip on the airplane together, and he said, "You who later became Head of Bioengineering at Western know, you really ought to make a department so your faculty, and he headed up that group that was at the their training had been." They were getting degrees 34 Engineering School. Most of our activity at the time in engineering, in all kinds of things that really didn't was still over at LDS Hospital, even though we were describe what they were trained in. And so, he brought part of the University all this time. together the Dean of So, Sam set that Engineering and the Dean up, and that was of Medicine and me, and the beginning of we talked about it. They bioengineering there. A asked me what I'd prefer. few years later, the Dean I said, "Well, I'd really of Medicine changed, prefer Medicine. That's and the new Dean felt my background, and I'd differently about us. be happy to be in the He felt that we really College of Engineering, should be located over at if that's the way it works the medical school. So Biophysics, 1975; (back row L-R) Mark Scolnick, John out." Morgan, Hyrum Marshall, Peter Barber, Justin Clark, Reed we did that. We broke And that's what Gardner; (front row L-R) Alan Toronto, Marion Dickman, Paul up the Department into Clayton, Alan Pryor, Curtis Johnson, Homer Warner. worked out. Turned out the Bioengineering that the Dean of the Department in the College of Medicine said, "There's no precedent for Engineering School, and the Department of Medical a department that doesn't train MDs-having a pure Biophysics and Computing (which soon became PhD kind of training program is outside the tradition the Department of Medical Informatics) moved of our medical school." But the Engineering School administratively to the School of Medicine. We didn't was tickled about it. Bioengineering was something move physically, we moved administratively, and we that had sort of caught on around the country at the had a presence in the Medical School. So that was time. nice. We formed the Department of Biophysics and And then it really wasn't until the University Bioengineering and were supplied space in the of Utah Hospital became a customer for the HELP Engineering building. We set up so we could do system that we actually moved faculty to the Medical animal experiments over there, big dogs housed and School. When the HELP system began to mature and experimented on in the Engineering School-it was we got a lot of attention, both from NIH and from quite a breakthrough. W. Sanford [Sam] Topham, other institutions, it became apparent that we ought Dr. Warner, ca. 1970s, in his laboratory coat, leaning on a Homer on his sailboat. Control Data Corporation (CDC) 3170 computer system. 10 to have some way of distributing this to other people. year after the sale, before you really got it installed, Then, NIH took it upon themselves to fund building everybody trained, and any income came from the two HELP-like systems: one at Massachusetts General installation. It was a drain on cash, and the company and the other at George Washington University. Then, didn't have the cash. So to make a long story short, we encouraged some of our local people, including Control Data, who was one of the big investors in my brother [Rick Warner], to start a company here in the company and who made the computer we used, Salt Lake City to supply such systems commercially. I bought out the other investors at a very cheap price, unfortunately. So the local investors didn't do very for some of our graduates. We got the company, called well. But Control Data bought out MedLab. Then they MedLab, up and going. None of the faculty owned any made a successful business out of it. The system has stock in the company. We thought that having the LDS been installed since then, and eventually 3M bought it. Hospital own part of the company might be a way of But there have been copies of that in a lot of hospitals the hospital getting some returns on their investment around the country, and even in Germany. Remember, and their support over the years. Al Pryor and Reed Gardner took a three-month So that company, largely funded by local investors, sabbatical and went over to Kiel in Northern Germany got off to a good start. They sold a number of systems, to help install one of the early HELP systems in a but every time they'd sell a system, they got further children's hospital. in debt, because you had to have the system about a Homer Warner, Al Pryor, Reed Gardner in the computer room at Intermountain Health. HELP was expanded throughout much of LDS Hospital over the following two decades, after its launch in 1967. While decision support remained the focus, automation May 16, 2005 11 DS Yeah, one of my friends got his PhD at that of years. But then some of the journals got upset, hospital, Ulrich Prokosch. particularly the Journal of Biological Chemistry. They got upset to the point where they refused to publish HW Oh, yes! anything that had been circulated before it was sent to DS Then Ulie ended up coming back over to Utah them. And so they shut us down, essentially. as well. So now he's a chief When that happened, Bruce of the Informatics Department I suspect I've D. Waxman said, "We need a journal. Would you be willing at the University of Erlangen. visited most of to edit a journal?" And so we the institutions But he was at the Institute did. We worked out a deal with for Medizin Informatik at the Academic Press and started the University of Giessen Hospital in Germany, where they that were doing journal in March 1967. I called on 15 of my colleagues around had the HELP system, with anything in the country who were leaders Professor Jockheim Dudeck. informatics around Editorial Board. So I edited the HW That's right. Nice people. the country. journal for 26 years, and it was DS I'll say. Very good friends, a tremendous opportunity. yeah. So the other thing I was I look back at my career, interested in was, when you founded the journal and I've really been blessed. You know, to get selected Computers and Biomedical Research. That was an to be on an NIH study section is a wonderful thing, important event. particularly in those days, because we made site visits, HW At the time, I was Chairman of the NIH had the opportunity to talk to people who were doing Computer Research Study Section. NIH had instituted research-not about what they had done, but what a program in enzymology, where the advances were they were thinking about, what they were going to going so rapidly that they thought the information do, and why they were doing it. You asked questions, ought to be circulated much faster than it is in ordinary and you go into an institution like that, and tell them journals. And they had what right off the bat, "Look, we are they called the Information We'd have a lot going back to a study section, and the two or three of us are Exchange Group, among these enzymologists. Bruce of fun together, a going to have to represent D. Waxman, who was the you. We are going to have Executive Secretary of this to sell your project to these study section with computers, fertilization, I other people, so we have to ask all kinds of things." They suggested that we do the same thing. So I did. I started this think, amongst our understand that. Talk about information exchange group, graduate students. an educational experience! I mean, it was just wonderful. and NIH paid all the expenses for it. What we would do is, I suspect I've visited most of when you had a paper that was ready to submit for the institutions that were doing publication, you would then submit a copy to me. I'd anything in informatics around the country, as a result send it to NIH. NIH would duplicate it and send it out of that wonderful opportunity. And a lot of the people to everybody who was on the list. Now, to get on the list, all you had to do was submit one paper yourself. was wonderful. Here are these non-reviewed papers getting And of course editing the journal was another kind circulated, and the only risk was to your reputation. of thing. I'd get the articles before anybody else. I had If something wasn't right, it was you that suffered. the Editorial Board that I'd send them to, for review. They would give me their reviews, but ultimately, there would have to be somebody who said yes or no. I 12 that I really understood what they were doing and pass DS I'll say. He said the same thing about you. He some kind of judgment on the manuscript. was telling us stories about when you'd go to the So all those things just made it a very special time meetings and then go to church together: go to the for me. I look back, and I'm just grateful for that point Mormon Church and then the Quaker Church. in time when all those opportunities came along. It's HW We did that, that's right. been a wonderful thing. DS The other thing that I We decided to DS I like that. was thinking about that was interesting was, we talked try to develop a HW We'd always go to the- oh, golly, what's the name of with Octo [Barnett]* earlier, teaching program. the meeting before it became-. and he was telling about how he met you, and you coming That's how we got DS out to some of those study sites. Can you tell us a little into Iliad. HW SCAMC, yeah. We'd go to SCAMC, and we'd walk bit about your relationship over to a place where they had crab, where you can eat all the crab you want to. It's HW Octo is one of my best friends. We've really really a fun time. He'd bring his graduate students, and I'd bring mine, and we'd have a lot of fun together, really got acquainted with Octo was when we made a lot of good cross-fertilization, I think, amongst our a site visit to Massachusetts General Hospital. Octo graduate students. Did he explain that we had Iliad was a junior faculty person there. They were debating [Homer Warner's diagnostic program]? We did a lot of whether to move ahead the same things along the with computerization, way. And he really came and what kind of person from a cardiovascular kind to get to head up this of background, too, you effort and so on. After know. So we had a lot in the site visit, I felt so common. positive about Octo and DS So, speaking of how bright he was, and Iliad, I remember when what his potential was, I got to LDS Hospital, that I wrote a letter to you left the Hospital the Administrator of the hospital-a prominent over to the University of guy, I can't think of his Utah Medical Center and name right now-but really started working on anyway, recommending that Iliad program. Can Octo. Of course he got Homer Warner, Jr. recalls, "For many years we took an the job, and he's been in annual vacation to Laguna Beach and stayed in the same rental home there. Happy memories." HW Well, the reason I that job for a long time moved from LDS Hospital and made tremendous to the University was because the University bought contributions. the HELP system. My friend Chase N. Peterson, MD, DS Absolutely. was President of the University of Utah, and he wrote me a note that said, "Well, we bought your system, HW Wonderful friend. We've had a lot of fun what are you going to do about it?" In other words, together. He's been out to our place in St. George, and "Aren't you going to come over and help?" we've sailed to Victoria together on our boat. We've done a lot of fun things together. Yeah, he's a special DS Make sure it works. guy. May 16, 2005 13 HW Yeah. I thought I'd send Al or Reed or one of the happy customers that bought our HELP system; it younger faculty members over there, but they said, didn't work out. And that's how I got into doing other "Oh, no, it's mainly a political kind of challenge, and things in informatics. you'd better go." So I gave up my career research Bill [William D.] Odell, who was Chairman of the award from NIH, and went over there with the intent Department of Medicine, asked me one day if there of adapting that system to their was something we could do needs. But that was one of the frustrations of my life, because And after a while, with computers that would facilitate the medical clerkship. I got over there, and the Dean when they begin He said that the lectures they'd been given to supplement the to see the system of the College of Medicine was very good; he bought us a experience of the student Heads computer just like the one the performing like on wards weren't well attended, LDS Hospital had-smaller, and the faculty didn't like but the same machine. We they perform, they them, either. And so that was their specialties were and make realize that it's a special opportunity for us. We decided to try to develop a whatever changes. But the their knowledge teaching program. That's how fellow who was in charge of the IT for the hospital convinced that's making we got into Iliad.35–39 It was a special set of circumstances, the Hospital Administrator that the system run, because there was a company called ALPS, Automatic and they become if he let us get into the software, we were bound to ruin it, and Language Processing. It's they'd lose their warranty with excited about it. technology out of BYU Control Data, who produced it. [Brigham Young University], facilitated language translation DS Oh, goodness. for industry. My brother Rick [Richard L. Warner] had HW So even though I had several good programmers that really knew the system very well, we were owned the largest Ford Dealership in Utah. unable to do anything with that. And as a result, they DS Yeah, I remember. He's the successful one in only implemented the business part of the computer. your family-is that the way your mother thinks of They didn't really ever get into the clinical stuff. Unfortunately, the University became one of the least Drs. Homer R. Warner and Stan Huff, now Chief Medical Homer R. Warner, 2011. dedication and opening night of the Intermountain Healthcare Homer R. Warner Center for Informatics Research, in 2011. HW That's right. [laughter] No, I don't think that's sat in on those knowledge engineering sessions-that the way Mother thought of it. She rather liked doctors would have been ideal. But at any rate, over this period of maybe six or company, and on the Board of Directors, they had eight years, we built the Iliad system, paying these a fellow, Ray Noorda-he was the CEO of the big consultants from the clinical specialties $50 an hour, networking company Novell. which wasn't very much, but He was a very capable guy and a very wealthy man. He There aren't very it was enough to keep them coming. Eventually, they got was on the Board of Directors many people who so interested because it was their system. It's really very are in academic of ALPS, and was working with my brother. I told Richard interesting: You start off saying, that we couldn't do this medicine whose "Gee, I don't know anything unless we got some people to about computers, what am I support us, because we had goal is to be a doing here?" And after a while, to have somebody to pay the consultants. We had these one- good teacher and when they begin to see the system performing like they hour sessions of knowledge care about the perform, they realize that it's engineering once or twice a week, with each of about entire diagnostic their knowledge that's making the system run, and they 15 different subspecialists, process. become excited about it. So the very best people of the that's how we got into Iliad. Medical School. It was a So ALPS originally put wonderful thing to bring those people in and get up some money, and then Ray Noorda from Novell them to interact with computers. Another educational decided he wanted to buy Iliad. So he did. He actually opportunity occurred for me to better understand hired my son, Homer Jr., to run that company, the little the broad scope of medical practice by having these company that he built. He had a half a dozen of these little companies that he just spawned off. I only met never really been a practicing clinician. Here I had the guy twice in my life. We didn't ever have anything an opportunity to talk with these experts in each of to do with him. But he had people that would provide these subspecialties about how they made a diagnosis. Unfortunately, that would have been the best thing for the medical students to do, too, so they could have all Homer R. Warner at the podium at Intermountain Healthcare. Homer Warner, Jr. with Jeanne and Homer Warner. May 16, 2005 15 don't we have this?" But unless you've got somebody grant from NIH. whose career is linked some way to development of Once we got it up to where we could begin using such systems, who can say, "This is good for me," it it, we integrated it into the teaching for the medical doesn't happen. It's a strange business. There aren't students. During their 12 weeks on Medicine, they very many people who are in academic medicine were required to use Iliad at the bedside to solve whose goal is to be a good teacher and care, in a their real diagnostic problems with real patients, and generic way, about the entire diagnostic process. also to solve two teaching and one testing case and a Instead, they are more concerned about their future simulated case on Iliad. Now, those simulated cases and how many publications they can get, rather than were really real cases that we'd developing diagnostic tools generated. We could make them We did some of like Iliad. See what I mean? It's an interesting business; you to what the professor wanted. these experiments wonder what's going to happen. We tried to evaluate it. It's on ourselves. In the future, somebody will come along and reinvent Iliad. because you say, it's better But they sold it. Soon than what? You start trying after I left, Homer Jr. left that company, too, and went to work for another company, we really can't tell whether this experience using a a laboratory company. Bayesian approach is going to teach the student to be a better diagnostician, but we can tell whether DS Oh yes, Sunquest. the student learned something about diagnosing a HW Sunquest. He worked for Sunquest for a number particular disease that he's seeing with Iliad, better of years, and then went to 3M. So he left, and they than he did with something he hasn't seen in a given sold the rights to market Iliad to a big publishing specialty. So, unbeknownst to the students, two company. The University still owns it. While we were different groups of students would be assigned to in Europe, I heard about it. But within a matter of maybe three subspecialties here, and the other group to another set of subspecialties they were being trained on. There were so many cases, they didn't really care it hasn't been maintained. It sort of breaks my heart, or know-it didn't make a difference to their training. because a lot of effort went into that. It was a lot of Then, we'd test them on cases in the specialties they'd fun. You'd like to see something that works there that been trained in, against the specialties they hadn't been you could point to. trained in. With that kind of approach, we were able to DS But unfortunately, again-it's an interesting lesson, Dean, what happens. When I retired, this was in full HW I'm taking all your time. operation. Mike Lincoln was running the program. He'd train new students when they'd come on how to DS No, I'm okay. I had another question. You use it. Very straightforward. When I retired, William mentioned the SCAMC [Symposium on Computer D. Odell, the Chairman of Medicine, retired as well. I Applications in Medical Care] meeting, and I went to Europe on a mission for the [Mormon] Church know you were involved with helping to form the of Jesus Christ of Latter Day Saints, and the Chair of AMIA [American Medical Informatics Association] Medicine did not even know that Iliad existed, and organization and ACMI [American College of Iliad was dropped just like that. It's very interesting, Medical Informatics] and everything. That's one because now I teach a course, an elective for senior of Don Lindberg's* interests, recording what happened. medical students in Medical Informatics, just to give HW Well, as I remember it, I was on the Board along on, and Iliad's part of that course content. We show them how to use Iliad in one of the sessions. They can't believe it and say, "Why aren't we using it? Why DS Yeah, there's a great picture of all of you guys. 16 HW We were not that unhappy with the way things HW It was too bad, but he felt very strongly about it. were going-I wasn't, anyway. I'm not much of an But yeah, I remember the meeting. I was not a leader organization guy. But it's clear that there were those in that sense, making it happen, by any means. I would that did feel we needed to do something in order to say Octo and Don were the key people that did that, bring together several groups so they could focus on made that happen. one group. They also wanted to build a journal that DS I was going to ask you what you were most had the name Medical Informatics. I'm really not sure proud of in your career, and you mentioned your where the name Medical Informatics came from. I peripheral vascular resistance controlling the can't trace that down. I think Europeans used it before cardiac output. Let's say that's one part of your we did. But anyway, they wanted to have a journal, they wanted to have a group that had a formal process. I think ACMI was an HW I don't think afterthought. I don't anybody would even think that happened at the know that I did that. same time. But Helmuth DS Well, I know, that's Orthner, you know, was a ways ago. really at the grassroots of SCAMC, one of the HW When you get to be chief players there. Octo, my age, well, how are really, I would say, took they going to remember the leadership role in Grandpa when he's making this thing real. In gone? [chuckles] I don't many of the discussions, know, I guess mostly I he really dominated and remember how lucky I did a great job at sort of have been and how many directing it. opportunities I've had. I One time, always look forward to Bill [William S.] going to work. It's just Yamamoto-do you exciting to see a new know him? remember when we did DS Yeah, I think I met him once. Homer R. Warner with the mold use to cast his sculpture. artery transfer function. HW He was at Alan F. Toronto, who the University of was working with me at Pennsylvania, but he was very much opposed to it. the time, and I were really questioning whether we He was going to sue! He thought we were doing ought to publish this. We thought we had our hands on something illegal, that some way, this wasn't a legal thing to do. really thought, "Let's do some more work before we publish this, because we don't want everybody to hear DS Was he the head of another one of the about it before it's ready." Of course [laughs] nobody ever did, anyway. But we were that excited about it. HW He was the head of SCAMC at that time. And We thought, "This is a whole new world opened up he'd been involved in SCAMC for quite a while. I to us," because I am not a mathematician, you know. can't remember all the details-but I do remember it I love mathematics; I'm not really good at it but I was sort of a bad thing, because his feelings were hurt enjoy it. But here you can solve differential equations by it. It was not a pleasant thing for any of us. without having to solve them analytically. All of a sudden, a whole world just opens up to you. DS I can imagine. Now we can go in and do things quantitatively- before, people just observed things. Dr. Heymans in May 16, 2005 17 Belgium received the Nobel Prize for discovering slow the heart down, but you could speed it up. So we the carotid sinus. Well, all he found was that as the were interested in the question, does increasing the arterial pressure increased, there was an increase in heart rate affect cardiac output? In a normal guy like you, as you exercise, your stroke volume stays about sinus. Our contribution was to develop a dynamic the same and your heart rate will double. So we tried and quantitative model that described the nerve that. While we were at rest we put the catheter into the atrium and increased the heart rate. We then measured developed a model for the sino auricular [SA] node in the cardiac output at different heart rates. We found the heart. Some years later, we found out that we were that we could change the heart rate from 60 up to 120, and the cardiac output stayed exactly the same. Isn't We did some of these experiments on ourselves. We that interesting?40 made a catheter with a wire down the middle of it and a piece of solder on the end. We inserted that catheter DS in ourselves, and we put it down into the right atrium, HW Yeah, it's very interesting. up against the wall of the atrium. Then, we could stimulate the atrium and drive the heart to whatever DS So I guess the stroke volume just goes down, rate you wanted within certain limits. You couldn't Homer R. Warner with the sculpture cast in his honor for the dedication of the Intermountain Healthcare Homer R. Warner 18 HW Oh, yeah. Well, the thing that determines cardiac HW The fellow down there, who was in charge, was output is the peripheral resistance. I've got heart a very famous mathematician, Robert E. Bellman. failure myself right now, and I take a beta blocker. He developed what we called dynamic programming. Well, what a beta blocker does is virtually dilate He was a very bright guy. He invited us down to the arteries. That's exactly talk about what we were what I was doing, taking a load off the heart, you see. The models that doing, because we were all doing quantitative things in The cardiovascular system we build with a physiology. They had these is amazing, because it is a pressure-regulated system. computer, the physicists and mathematicians that had been so valuable models we build during the war, and they wanted in informatics, the system. What happens is, the metabolism in the muscle, for something else now. So as soon as you exercise, as are useful we went down, and we spent soon as that muscle contracts a week down there-just the once, the CO2, the lactic acid, because they're three of us. builds up, and capillaries dilate. No question about it. August quantitative. We One night, we were over at this mathematician's home, Krogh, [Nobel Laureate 1920] can do things you and I raised this question with one of the old physiologists, a Danish fellow, showed that can't do with the him. I said, "One of the things I'm worried about, I've got principal in frogs, way back unaided human these models that we build, and mind. around 1920. If you have the accept the data and so on. How two hind legs of a frog, and you do you know whether a model's stimulate the motor nerve to unique?" He said, "There's no such thing as a unique model." "Well, how do you and quickly freeze it, then look at the frozen cross- judge a model, how good it is?" "Well, it's strictly sections under a microscope, the exercising muscle on aesthetic grounds." He then brought some of the will have over 100 times more capillaries open than physicists in to talk about different theories about life the non-exercising leg. Can you imagine that? Now and so on. You judge the model based on whether obviously, they're not the same hundred capillaries all the time. The capillaries in the tissue are opening relationship with your reality, is it useful, and so on. and closing all the time. It's a Well, that was an eye-opener dynamic thing. It's an amazing system. Morrie Collen had for me. I quit worrying about that. It's true, models are gross DS Amazing how this works, us-no question world, but they're very useful. The models that we build with HW Oh, it is amazing. about that. a computer, the models we Anyway, that fascinated me. build in informatics, are useful In informatics, I used to worry because they're quantitative. about whether the models that we built were unique. We can do things you can't do with the unaided human Along about 1970 or so, I was invited, along with Earl mind. You can't really do those things without a Wood and another physiologist whose name I can't computer. The concept, for instance, of stroke volume think of from the University of Pennsylvania, to the was useless until we got the computer. All of a sudden now, we're using the computational capabilities provided by the computer as the central focus for our DS Yeah. entire patient monitoring. However, we quit using the pressure pulse method to calculate stroke volume. Do you know why? May 16, 2005 19 DS I think so. HW He'd done the same kind of thing, only not nearly as sophisticated as we then made ours. We had a lot of HW You remember Rob [Robert] M. Cundick, Jr.? physiological measurements. We had blood pressure DS No, I don't know him. measurements, and we had forced vital capacity spirometry, and ECGs.41 All that was automated and HW He was a graduate student in our Department. We computer driven, and we had an automated patient used the pressure pulse method for years as the basis history.42 What Dr. Collen had was looking at the same for our cardiac output measurements. We'd calibrate kind of thing, only his was all done with punch-card it against the dye dilution curve measure of cardiac output. Then, we'd measure of thing. But it was his papers the changes in the patient. But we found that some of the It is the people that gave us the idea of doing multiphasic screening. That was patients, about 1 in 20, would that make the very successful, up to the point drift off, and it would give you a bad index as to which way the difference. where the payment system for medicine changed. The hospital cardiac output was changing. could no longer charge for that. Rob Cundick did a PhD dissertation on the topic, trying to reproduce what I had found in animals and the hospital, so they stopped it. It wasn't particularly in normal subjects, but he never could. The method because it wasn't good for the patient, because we was unreliable enough that we quit using it. It was discovered a lot of problems that made them postpone unfortunate, and it's still a challenge to me to know surgery, because they found these patients had a why on earth things changed. problem they hadn't recognized earlier. But, yeah, DS I heard Reed say one time that he spent a long time chasing that problem with I look at this little on us-no question about that. And I think Ted Shortliffe you. He said something like, it machine I've got remember him inviting me seemed like the capacitance or the stiffness of the heart was in my pocket, and down to Stanford to talk about changing somehow, and you I've got a thousand the HELP system once. I'd couldn't tell when that was always been sort of a critic of happening. times more HW Yes, that's right. We just memory in this whole notion of it. I think it's far-fetched. None of that stuff could never do it. Rob worked on that for several years, and than we had in that really came to fruition, but it there really was no good answer machine that ran was stimulating, an idea that was very interesting. I think for it. So, you win some and you lose some. the whole hospital Ted's done some good things, even though they're more on DS You mentioned Octo information system the theoretical level, and we've Barnett and the Ledley and 16 years ago. been over on the practical end. Lusted paper24 and some It takes both. other people. Were there other people who had big effects on your life and your DS I know there was some tension when I was a graduate student, it seemed like, between the groups at Stanford and Utah. HW Well, Morrie [Morris F.] Collen* did. I don't I know Reed Gardner one time said that Ted even remember-probably before your time-but we Shortliffe said, "Well, you guys are just a bunch of instituted an admission screening laboratory at LDS engineers!" He meant it as sort of a put-down, and Hospital, inspired by his work. Reed said he took that as the biggest compliment he DS He'd done that same thing at Kaiser, I think. 20 had ever gotten. So I guess everyone was happy, at But that's what life's about-it's about the journey, it's least. not about the end product, I don't think. HW [laughs] That's right. DS I was going to ask you if you had any words of wisdom for people that are new, coming into the DS You mentioned one of your biggest disappointments was that the HELP system didn't you another chance. work at the University of Utah Hospital. Would you HW I look at this little machine I've got in my pocket, and I've got a thousand times HW Yeah, that was one more memory in this than we of them. You know, you had in that machine that ran learn a lesson from all your the whole hospital information experiences. The lesson is that system 16 years ago. It's a "it is the people that make thousand times faster, and it the difference." You know, costs a thousand times less. You you've got a new tool that may know, that's pretty exciting. look great to you, but unless Now, how can anybody not somebody out there thinks be enthused about where this it's in their best interest to use it, make money off of it, or got to think about it, think whatever the motive might about the Internet and all the be, it's not going to survive. things that are happening, and That's a little disappointing. the impact it's had. So, the That's not an idealistic way to opportunities for doing things go, but it works. It's the free are greater than they have enterprise system, in a sense. ever been. It's a tremendous I guess that's why I fall back, 43 I really think that-and in a sense, on our modeling in I hate to keep referring back the physiological domain as a to the modeling-but I really more rewarding kind of thing think that's going to be the big in the long haul-because, contribution that computers are you know, somebody else is Portrait of Dr. Homer R. Warner, painted by going to make. It's going to internationally acclaimed artist Galina Perova. going to improve on it. And allow us to build models to help us understand the world around Everybody's working on it, and you make a little us, and take advantage of these resources that we have. contribution here, and somebody else makes it better there. You know, you look at Randy [Randolph] A. DS I think people are doing that. Miller, for instance, what happened to QMR? [Quick HW Oh, I do too. Look what's happening Medical Reference44,45-Reed Gardner noted in 2015 in imaging-just an example of it. That has that QMR inspired Homer Warner to create Iliad.] He revolutionized medicine. It's had a tremendous impact. leaves Pittsburgh, and it's dead. It's the same story, Guys like Dennis L. Parker are doing a tremendous isn't it? job. I don't know where it's all going to go. I guess if DS Yup. I had a wish, it would be to be 30 years younger and able to see what's going to happen. HW You think of the years that went into the QMR effort. It was a wonderful thing, stimulating, DS everybody was interested in it, and it provided a lot of HW It's going to be exciting. It's tremendous. intellectual fodder for the whole community and the DS You know David M. Eddy, he's a physician, and he works with Kaiser Health Care a lot. He's built May 16, 2005 21 some huge models of diabetic care that model all the HW Well, it was nice to visit with you again. way down from the cellular level of diabetes, all the way up to the institutional level. And so he models DS Yeah, really good to see you. I'm glad you're thousands of patients who have heart problems and doing well. Reed says you're still going to the they have diabetes, and then they come to clinics, seminar, and teaching a class, and other things. HW Well, I get up there a little bit. You've got to He's made some tremendous discoveries. He's have something to keep you stimulated. I've been been able to duplicate some of these huge trials working with Alan H. Morris a little bit. He's been that they've run, where they run a 10-year clinical working with these protocols to manage clinical trials. trial with 40,000 patients. They cost something like He's now got a nice grant, a million-dollar grant to $150 million, and he shows that putting the same disseminate these "tools," we call them, that really input conditions into his model, he draws the same Dean Sorenson has built, and he's been testing and survival curves as these studies do. He's getting a using for years. But disseminate that to these 25 lot of attention now with this kind of a model, and he's trying to build it in some other diseases to try to institutions and show that they really can facilitate the understand these diseases. clinical trials, you know. They control all the variables that otherwise are just random, and you have to have a HW He's an interesting guy. I don't know him lot more patients to get the same result. So that's been personally, but I remember we had him as the fun. That's about it. keynote speaker once at the SCAMC meeting. Very stimulating. DS Well, this has just been wonderful, and I certainly thank you, and everyone else will thank DS Yeah. He reminds me a lot of you-he's a you as well. mathematician and a doctor as well, just like you, building these models. HW Very interesting guy. [Published 16 April 2015; updated , 2015] DS Well, I can't thank you enough. This has just been wonderful. 22 23 For additional Conversations with Medical Informatics Pioneers, please visit: http://lhncbc.nlm.nih.gov/project/medical-informatics-pioneers References Ash JS, Sittig DF, interviewers. Conversations with Medical 11. American Medical Informatics Association (AMIA). Morris * F. Collen Award of Excellence Recipients - Video Tributes. Informatics Pioneers. Goodwin RM, Ash JS, Sittig DF, editors. Bethesda: U.S. National Library of Medicine; Available at: http://www.amia.org/programs/acmi-fellowship. 2015. Available from: http://lhncbc.nlm.nih.gov/project/ medicalinformatics-pioneers 12. Warner HR. The role of computers in medical research. 1. Patton GA, Gardner RM. Medical informatics education: the University of Utah experience. J Am http://www.pubmedcentral.nih.gov/articlerender. 13. Budkin A, Warner HR. Computer assisted teaching of fcgi?artid=61389&tool=pmcentrez&rendertype=abstract. 50. Available at: http://www.ncbi.nlm.nih.gov/ pubmed/5721567. Accessed April 7, 2015. 2. Warner HR, Olmsted CM, Rutherford BD. HELP--a program for medical decision-making. Comput Biomed Res. for use in healthcare locations. 2008. Available at: http:// www.google.com/patents/US7315825. Accessed April 7, 2015. 3. Gardner RM, West BJ, Pryor TA, et al. Computer-based ICU data acquisition as an aid to clinical decision-making. 15. Homer Warner, MD, PhD - Department of Biomedical Crit Care Med. 1982;10(12):823-30. Available at: http:// Informatics. Univ Utah - Sch Med - Salt Lake City, Utah. Available at: http://medicine.utah.edu/bmi/scholarship/ 2015. warner/. Accessed April 6, 2015. 16. Weber B. Homer R. Warner, a Pioneer of Using Computers system. J Med Syst. 1983;7(2):87-102. Available at: http:// in Patient Care, Dies at 90 - NYTimes.com. New York www.ncbi.nlm.nih.gov/pubmed/6688267. Accessed April 7, Times. http://www.nytimes.com/2012/12/11/us/homer-r- 2015. warner-a-pioneer-of-using-computers-in-patient-care-dies- at-90.html?src=recg&_r=3&. Published December 12, 5. Pryor TA. The HELP medical record system. MD Comput. 1988;5(5):22-33. Available at: http://www.ncbi.nlm.nih.gov/ pubmed/3231033. Accessed March 12, 2015. 17. Robinson D. Utahn Homer Warner led a full life of service to all. Deseret News. http://www.deseretnews.com/ 6. Kuperman GJ, Gardner RM, Pryor TA. HELP: A Dynamic Hospital Information System. New York: Springer-Verlag; service-to-all.html. Published December 3, 2012. Accessed 1991. 7. Haug PJ, Gardner RM, Tate KE, et al. Decision support 18. Owens-Liston P. Homer Warner, Founder of Biomedical in medicine: examples from the HELP system. Comput Informatics. University of Utah Health Care Health Feed: Expert Health News & Information. http://healthcare.utah. ncbi.nlm.nih.gov/pubmed/7813202. Accessed April 7, 2015. edu/healthfeed/postings/2012/11/113012homeobit.php. Published November 30, 2012. Accessed September 10, 8. Warner HR. Computer-Assisted Medical Decision-Making. Imprint New York: Academic Press; 1979. 19. Lee VS. Homer Warner -- Founder of Biomedical 9. Warner HR, Sorenson DK, Bouhaddou O. Knowledge Informatics [Video]. Univ Utah Heal Sci Blog. 2012. Engineering in Health Informatics. New York: Springer; Available at: http://healthsciences.utah.edu/notes/postings/ 1997:262. Available at: http://www.weyrich.com/book_ november_2012/113012HomerWarner.php#.VBCnnleGc-I. reviews/knowledge_engineering_health.html. Accessed April 7, 2015. 10. Clayton PD. Presentation of the Morris F. Collen Award to Homer R. Warner, MD, PhD: "why not? Let's do it!". at: http://www.pubmedcentral.nih.gov/articlerender. 21. Warner HR, Toronto AF. Effect of heart rate on aortic gov/pubmed/13783191. Accessed April 7, 2015. May 16, 2005 22. Warner HR, Russell RO. Effect of combined sympathetic Symp Comput Appl Med Care. 1991:1005-10. Available and vagal stimulation on heart rate in the dog. Circ Res. at: http://www.pubmedcentral.nih.gov/articlerender. gov/pubmed/5780152. Accessed April 7, 2015. Accessed April 7, 2015. 36. Guo D, Lincoln MJ, Haug PJ, Turner CW, Warner HR. output from dye dilution curves recorded by oximeter. J Exploring a new best information algorithm for Iliad. Proc Appl Physiol. 1952;5(3):111-6. Available at: http://www.ncbi. nlm.nih.gov/pubmed/12990551. Accessed April 7, 2015. at: http://www.pubmedcentral.nih.gov/articlerender. Accessed April 7, 2015. Diagnosis. Science (80- ). 1959;130(3366):9-21. 37. Lincoln MJ, Turner CW, Haug PJ, et al. Iliad training 25. Warner HR, Toronto AF, Veasy LG. Experience with Bayes' enhances medical students' diagnostic skills. J Med Syst. Theorem for Computer Diagnosis of Congenital Heart 1991;15(1):93-110. Available at: http://www.ncbi.nlm.nih. April 7, 2015. 38. Lincoln MJ, Turner CW, Haug PJ, et al. Iliad's role in the generalization of learning across a medical domain. Proc 26. Warner HR, Toronto AF, Veasey LG, Stephenson R. A mathematical approach to medical diagnosis. Application to at: http://www.pubmedcentral.nih.gov/articlerender. congenital heart disease. JAMA. 1961;177:177-183. Accessed April 7, 2015. 27. Warner HR, Toronto AF, Veasy LG, Stephenson R. A mathematical approach to medical diagnosis: Application 39. Guo D, Lincoln MJ, Haug PJ, Turner CW, Warner to congenital heart disease. JAMA. 1961;177:177- HR. Comparison of different information content 83. Available at: http://www.ncbi.nlm.nih.gov/ models by using two strategies: development of the best information algorithm for Iliad. Proc Annu Symp 28. Warner HR. Editorial: Evolving a computer facility. http://www.pubmedcentral.nih.gov/articlerender. Comput Biomed Res. 1967;1(2):i-ii. doi:http://dx.doi. Accessed April 7, 2015. 29. Warner HR, Gardner RM, Toronto AF. Computer-based monitoring of cardiovascular functions in postoperative of pressures in man by cardiac catheters. Circ Res. gov/pubmed/13172871. Accessed April 7, 2015. 30. Warner HR, Morgan JD. High-density medical data system for decision-making by computer. Am J Cardiol. management by computer. Comput Biomed Res. pubmed/622991. Accessed April 7, 2015. 31. Gardner RM. Monitoring of physiological data in a clinical Bayesean approach to history taking and diagnosis. Comput Biomed Res. 1972;5(3):256-62. Available at: http:// 2015. 32. Warner HR, Woolley FR, Kane RL. Computer assisted instruction for teaching clinical decision-making. Comput at: http://www.pubmedcentral.nih.gov/articlerender. fcgi?artid=116258&tool=pmcentrez&rendertype=abstract. 33. Haug PJ, Warner HR, Clayton PD, et al. A decision-driven Accessed March 6, 2015. system to collect the patient history. Comput Biomed Res. 1987;20(2):193-207. Available at: http://www.ncbi.nlm.nih. gov/pubmed/3595100. Accessed April 7, 2015. Reference Source Information," UMLS Source Release Documentation. Available at: http://www.nlm.nih.gov/ research/umls/sourcereleasedocs/current/QMR/ Available from: http://www.openclinical.org/aisp_qmr.html. 35. Lau LM, Warner HR. Performance of a diagnostic system (Iliad) as a tool for quality assurance. Proc Annu 25