Oral-History:Robert Mates

About Robert Mates

Dr. Robert Mates is a prominent researcher in the field of biomedical engineering. His areas of interest include cardiovascular modeling, control of cardiac output, coronary circulation, and rehabilitation of patients with neuromuscular diseases. Dr. Mates has been a member of the faculty of the University of Buffalo since he received his Ph.D. in Mechanical Engineering from Cornell University in 1963. He has served as Chair of the Department of Mechanical and Aerospace Engineering from 1967-70 and 1979-83, has founded the Center for Biomedical Engineering in 1989 (and continues to serve as its director), and is actively involved in the ASME, among other organizations.

This interview begins with a brief discussion of Dr. Mates’ background and his undergraduate education at the University of Rochester. It continues with an overview of his graduate work at Cornell University and his decision to join the faculty at the University of Buffalo. He discusses his early research in hypersonic aerodynamics, his work for NASA and the Air Force, and his decision to begin studies of heart disease and the physiology of the heart. Dr. Mates also describes his work with Fran Klocke and David Green, his use of instrumentation and models, the development of engineering science, his formation of the Center for Biomedical Engineering, and his work on the rehabilitation of neuromuscular diseases.

About the Interview

ROBERT MATES: An Interview Conducted by Frederik Nebeker, Center for the History of Electrical Engineering, 13 December 2000

Interview #278 for the Center for the History of Electrical Engineering, The Institute of Electrical and Electronics Engineering, Inc.

Copyright Statement

This manuscript is being made available for research purposes only. All literary rights in the manuscript, including the right to publish, are reserved to the IEEE History Center. No part of the manuscript may be quoted for publication without the written permission of the Director of IEEE History Center.

Request for permission to quote for publication should be addressed to the IEEE History Center Oral History Program, IEEE History Center, 445 Hoes Lane, Piscataway, NJ 08854 USA or ieee-history@ieee.org. It should include identification of the specific passages to be quoted, anticipated use of the passages, and identification of the user.

It is recommended that this oral history be cited as follows:

Robert Mates, an oral history conducted in 2000 by Rik Nebeker, IEEE History Center, Piscataway, NJ, USA.

Interview

Interview: Robert Mates

Interviewer: Rik Nebeker

Date: 13 December 2000

Place: La Jolla, C.A.

Childhood, family, and education

Nebeker:

I’d like to start by asking you where and when you were born, and a little bit about the family you came from.

Mates:

I was born in Buffalo, New York, on May 19, 1935. I have one brother who is six years younger than me, and we were the first generation of the family to go to college. Neither my mother or father had even a complete high school education. So that was unique. They were both very interested in education and encouraged us, but neither of them had the opportunity.

Nebeker:

What did your father do?

Mates:

He worked in an office. He was a manager of a grain milling office. He sort of worked his way up. He had actually immigrated from Ireland to Canada in the 1920s and somehow migrated to Buffalo.

Nebeker:

He was born in Ireland?

Mates:

Yes. In fact, my brother and I and our wives just came back from Ireland in September. We went to visit the town in which he grew up. But I can’t trace the history back any further than that. There are no relatives left over there now. I have five cousins who all grew up in Canada and were all children of one of his brothers. They are the only known relatives that we have.

Nebeker:

And your mother’s family?

Mates:

We don’t really know a whole lot about my mother’s family. She was born in Buffalo. She had a lot of relatives in Canada, but whether the family came from Canada I don’t really know. We never did have much history on her side of the family.

Nebeker:

When you were growing up, were you interested in science and gadgets and things?

Mates:

Yes. I took lots of things apart and rarely ever put them back together. I was always interested in mechanical things: bicycles and clocks and things.

Nebeker:

So many of the people that I talked to, these electrical engineers, were into crystal radio and amateur radio.

Mates:

One of the interesting things that I noticed as my career in teaching progressed was that all the kinds of things that we did as children, like taking apart and fixing bicycles, doesn’t happen anymore because everything is so complicated. I rebuilt entire engines when I was a teenager, and you can’t do that anymore. So none of my more recent students had any intuition about mechanics and how things worked. I think that’s a big drawback, and I don’t know how to solve the problem, but we certainly had much more of a head start in the mechanical areas. I imagine the same was true of the electrical.

Nebeker:

That’s true. With many of the electrical gadgets, people could build not only their own receiver but their own transmitter.

Mates:

Not anymore.

Nebeker:

Well, you buy a chip that does everything.

Mates:

One of the courses that I taught was the basic sophomore statics and dynamics, the mechanics course. It was amazing how little intuition people had about very simple levers and simple ideas that I sort of intuitively knew about.

Nebeker:

How did your school years go? Were you particularly interested in science?

Mates:

Yes. I was good in math and science, and I picked engineering primarily because I was interested in it. But it was the one area in which you could get a degree as an undergraduate that would earn you a living. That would be a good way to put it. My family had very limited means. I was fortunate to get scholarships to go to college, but I never, at that point, had any notion of anything beyond four years.

Undergraduate and graduate studies

Nebeker:

The University of Rochester is a private institution, right?

Mates:

Right.

Nebeker:

And you got a scholarship?

Mates:

I got a full scholarship. Actually, I was able to earn enough working to pay all my expenses. That, again, would not happen today. Now you end up with $20,000 or $30,000 worth of debt, which I didn’t have, fortunately. Actually, what happened there was that I got to be very good friends with a couple of the faculty. They obviously enjoyed what they were doing, and I liked the atmosphere, so I thought, “Well, let’s see if I can proceed further,” and I got a fellowship.

Nebeker:

With the idea of going into academia?

Mates:

Yes. That was why I went to graduate school. I went to Cornell initially on a fellowship. Then I became an instructor there and taught for three years. Again, that was an experience that you would probably not have today. They were very short on faculty in those days, so they hired graduate students to teach. I taught a full load, more hours as a graduate student than I ever taught as a faculty member, and I had my own sections of thermal and fluid mechanics and stuff.

Nebeker:

That was an education.

Mates:

That was an education. Yes. I liked it. I had some talent, at least, and so that’s what I did.

Nebeker:

Was it in ‘57 that you graduated from Rochester?

Mates:

In ‘57 I graduated from Rochester, and I started at the University of Buffalo in the fall of 1962, but I was writing my thesis at that point.

Nebeker:

What teachers were particularly influential? I certainly want to hear about your thesis advisor at Cornell, but was there somebody at Rochester?

Mates:

Actually, yes. There were two brothers, one at Rochester and one at Cornell. The Conta brothers. Louis Conta was at Rochester and was my primary mentor there. Probably the reason that I went to Cornell was that his brother Bart was on the faculty at Cornell. I liked both of them and that was the primary reason I chose Cornell. I mean, it’s an excellent school, but I could have chosen another school that would have been equally good. I would say that Lou Conta had the most influence on directing my career because he loved teaching and he was great at it. I thought that it would be a great way to spend my life.

Ph.D. thesis, Cornell University

Nebeker:

So once you decided on an academic career, how did you narrow your area of interest?

Mates:

Again, I suspect that was because of the faculty that I had. Conta taught thermo and fluid mechanics in that area, and he was the best teacher I had. So that was the area I found most interesting as an undergraduate. Then when I went to Cornell, I met my thesis advisor, Dennis Shepherd, who was interested in turbo machinery kinds of things. So I took some courses from him and did my Ph.D. thesis in the area of turbo machinery on a project that was sponsored by the Carrier Corporation. They made compressors and were still learning more about how the compressors worked.

Nebeker:

Was this a general thing or more specialized?

Mates:

The thesis was very specialized, but it was an experimental thesis. I liked doing experimental work, and we were able to get some sponsorship from the Carrier Corporation. This was in the days before government funding was really common, so much of the research that was done at Cornell in those days was funded industrially. There was one National Science Foundation grant in the whole department while I was there. So I did that and enjoyed that.

Nebeker:

I know so little about this. What sort of turbo machines? Where would I encounter them?

Mates:

The particular one that I worked on was what’s called a centrifugal compressor. It has veins on it, and as it spins it provides energy to the fluid and raises its pressure.

Nebeker:

Like a turbocharger in an engine.

Mates:

Yes. The particular one that I worked on was used in air conditioner compressors. But you would see turbo chargers on automobiles, and they use them for compressing natural gas in pipelines.

Nebeker:

And in your thesis work I assume you’d have a mathematical description.

Mates:

Well, at that point it was largely experimental. I did some correlations, but there was no sophisticated modeling involved because it was a very complicated problem. At the time I did it, the instrumentation used hotwire anemometry for measuring fluctuating flow fields. That was just coming into use and there was not a whole lot of theory of turbo and fluid flow at that point. It’s much more sophisticated now.

Nebeker:

So at that time, what was special was having good numerical data on this flow.

Mates:

Yes. Being able to describe things numerically. What we did to vary a configuration was that we took different designs of turbine wheels and measured the flow field at the outlet of those, and from that, tried to determine the efficiency.

Nebeker:

So it sounds like this was fairly practically oriented work.

Mates:

Yes. It was a very sort of down to earth kind of applications engineering. It wasn’t very sophisticated in terms of today’s world.

Nebeker:

How did you like the Cornell graduate school experience?

Mates:

It was an excellent school. I must say (and I have a chance to say it on the record here), it’s an awfully snobbish place, like most Ivy League schools. At the time I was there, they thought they were the greatest school in the world. They were good, but I don’t think they were as good as they thought they were, and they probably still think they’re great. Today I would say that they are a good school, but not at the very top of the heap. That was the thing I didn’t like, was that everybody was very self satisfied. When I told my advisor that I was going to the University of Buffalo he was horrified that I would take this great Cornell education and go out to the provinces.

Nebeker:

And “step down” to the University of Buffalo.

University of Buffalo

Mates:

It was a very pragmatic decision, actually. My mother was still alive and she was quite ill, and my brother was not in Buffalo. It was convenient for me to be near her. Also, it was a great opportunity because in 1962 the University of Buffalo became part of SUNY. So, it was a chance to get in on the ground floor and sort of influence the program much more than I could have.

Nebeker:

They already had, I assume, mechanical engineering.

Mates:

They had a program that had been there since the ‘40s, but it was strictly an undergraduate program. There was a very limited graduate program, largely part-time in the evening for engineers working in the industry. There was no Ph.D. program. The first year I was there, we got a Ph.D. program.

Nebeker:

So, you were brought in to build the department?

Mates:

I was the first of a lot of hires that came in, and in the first four or five years that I was there, the department probably doubled in size. That was in the Rockefeller years when money was no object. In fact, in a sense, my career was slowed down in a way. After I had been there five years, the department head retired. He was the founder of the school of engineering. He was a nice guy, but he had no aspirations. I was the oldest of the new guard, so they said, “Why don’t you become chairman of the department?”

Nebeker:

After five years?

Mates:

Yes, and this was crazy. But there wasn’t anybody else and we weren’t really in a position at that point to go outside for a search. So I said, “Well, I’ll do it for a year.” Then it went another year and another year. It was an amazing experience, though, because literally, money was no object. They would say, “We want you to go out and hire faculty.” “How many?” “As many as you can find.” They would call up on the phones and say, “We just found another $100,000 in equipment money, but it has to be spent by next Friday.” That kind of thing went on for several years, and while we certainly built a good department, it was not a very efficient way of doing business.

Nebeker:

So this is mid-‘60s?

Mates:

I became chair in the fall of ‘67 for the first time.

Nebeker:

Okay. So late ‘60s.

Mates:

And then in the ‘70s, New York went bankrupt and the roof came crashing down, but we had a few years of not worrying about the money.

Nebeker:

You must have been one of the youngest department heads in history.

Mates:

I probably was at that time.

Aeronautical engineering, hypersonic aerodynamics research

Nebeker:

How about your research when you arrived at Buffalo? What were you working on?

Mates:

Well, there was no graduate program to speak of. There were no graduate students to speak of. I was very fortunate there, and my whole career probably was a series of coincidences. The one coincidence there was that at that time Cornell had Cornell Aeronautical Laboratory in Buffalo. It was one of the top aeronautical engineering research groups in the world, and they had actually made me a job offer. When I decided to go to the university, they said, “Well, why don’t you come here during the summers, and you can work one day a week?” So I got my research career there. It was not started at the university at all. For probably the first five or six years, almost all of the research that I did was with that group at the Cornell aeronautical lab.

Nebeker:

As you know, Bert Fung was an aeronautical engineer, and as I think about it, your thesis work seems close to a lot of aeronautical engineering.

Mates:

As I said, I had taken a lot of aeronautical courses. That was where most of the theoretical fluid mechanics was taught at Cornell. I got a chance to apply a lot of that with the group at Cornell lab.

Nebeker:

What specifically was the research that you were doing?

Mates:

Hypersonic aerodynamics. I went from these little compressors to working mainly on the Apollo program.

Nebeker:

Okay, so this was contract work.

Mates:

It was largely NASA and some work for the Air Force, but the idea was trying to understand the re-entry physics of vehicles as they came back into the atmosphere.

Nebeker:

Maybe the big issue for the space program.

Mates:

Yes. They didn’t know how long this communications black out was going to be. So that was really where I got more into analytical and numerical work because what we were doing was simulating flow fields around the Apollo vehicle. It was big number crunching stuff. They had computer facilities.

Nebeker:

Do you remember on what computer they did these simulations?

Mates:

Well, the IBM 650 was the first one and then the 704, and there was a whole progression. But these were so numerically intensive that what we did was we rented the computers for the night shift. We would start our program at midnight, come back in the next morning and see where we had gotten over the eight hours. Then we would run the normal stuff during the day and we’d go back the next night again. Some of these must have taken 100 hours on the IBM 704 computer, but we were able to get some very good results. We were able to predict ionization levels in the plasma shield.

Nebeker:

And this was part of the Apollo Program that fed into design and other areas?

Mates:

Yes. We came fairly close to predicting the length of the communication blackout period, which was a big concern.

Nebeker:

I take it that those decisions had to be made largely on the basis of this modeling. Maybe you couldn’t do much in wind tunnels.

Mates:

They had shock tunnels that could produce very brief bursts of this kind of speed. They couldn’t simulate the whole flow field, but we used that to test some of our numbers, so it was a combination of that. It was largely analytical.

Nebeker:

How did your research evolve?

Mates:

Again, through coincidences.

Audio File

MP3 Audio
(278 - mates - clip 1.mp3)

When I became department chair, everything slowed down a lot. By that time, we had the Ph.D. program in place and we were starting to attract some good graduate students. One of the fortuitous things was that four of my first Ph.D. students were from the group at the Cornell lab. They would never have picked Buffalo out of the air, but they were there. Cornell would allow them to do a lot of their course work part-time, and would actually fund them for a year full-time to come over and do a thesis. So I had four really, really great Ph.D. students that came out of that association.

Cardiology modeling, NIH postdoc

Mates:

By then, the space program thing was starting to wind down a little bit, at least the basis research part of it, and so money was starting to dry up. That was probably in the early ‘70s. I was kind of looking for new things to do and again, fortuitously, I became very good friends with David Greene, a cardiologist at the university. We served together on the University Personnel Promotions Committee for three years. He was quite a bit older than I and was an intensely curious person, a very bright guy. But he had never studied anything about mathematics or physics. He would read articles in the cardiology journals that would have some equations. People were just starting to do some modeling and he would bring these to me and say, “Could you explain this to me?” I said, “Well, this is really kind of intriguing stuff.” It was getting harder and harder to get funding in the space areas, so I thought, “Well, let’s give it a try.”

So I took a sabbatical in the early ‘70s, and I went to the department of medicine at the university and got an NIH post-doc. I went over there for a year, went in the cath lab and watched them make measurements to see what sort of things they could do and couldn’t do. I sat in on a couple of courses to learn some of the basics. He [Greene] and I collaborated on a couple of studies, largely clinical studies at the beginning.

Nebeker:

What were these studies?

Mates:

It turned out in retrospect to not be as exciting as we thought it was at the time, but people were starting to calculate stresses in the heart, for example, using very simple models, and then trying to see if they were different in patients that had different kinds of heart disease. It turned out that they really weren’t as different as one might have expected. Then we also looked at the velocity with which the heart contracts. They could make measurements in the catherization lab of angiograms at twenty-five or thirty frames a second as the heart was contracting. You could get shortening velocity with the various fibers. So we did some studies with that.

Nebeker:

Was the modeling at that level, with the muscle fibers in the heart?

Mates:

The electron microscope was used in physiology for the first time in the ‘60s. Another sidelight was that an undergraduate fraternity brother of mine went to Cornell and we lived together for a couple of years before he got married. He was a solid state physicist who decided at the end of his Ph.D. that he wanted to get into biophysics. He went to England and did a post-doc with Andrew Huxley, of Nobel Prize fame, and he actually is the lead author on the first paper published showing an actual sarcomere with an electron microscope. I always thought that if Huxley hadn’t won the Nobel Prize for his neural work, that Al Gordon, my roommate, probably would have shared one for this work because that is still one of the most widely quoted papers.

Nebeker:

So this sarcomere wasn’t identified until the electron microscope?

Mates:

Well, they had never seen it. They knew it was there, but you couldn’t do it with optical microscopy. They were actually able to take pictures of it contracting. You could watch the film on slide. So anyway, at that point, when I was doing the work on the heart, we knew a little bit about muscle physiology, but not nearly as much as we do now. We didn’t really know the detailed structure of the ventricle. We knew that there were fibers running in different directions. Subsequently, there was a lot of work done in actually taking hearts apart and looking at the ways in which the fibers moved.

The result of all of this was that we were able to do some things. I got more intrigued with the physiology of the heart, and what we could do with patients was very limited because you could only make measurements of the overall behavior. During the time I was in the Department of Medicine, I became good friends with another cardiologist who was about my age, Fran Klocke, who was more of an experimental cardiologist. His interest was in the fundamentals of cardiac contractility and using animal models, where you could make much more detailed measurements. So I started working with him, and got into sort of a more fundamental level, not only in the muscle mechanics, but at that point we also got into modeling the behavior of coronary arteries, particularly coronary arteries that were partly constricted.

Nebeker:

Modeling the flow behavior?

Mates:

Yes. Modeling the flow behavior. We actually built a large coronary artery of plastic that we can instrument, and we had a pulsing pump that would simulate the flow of pressure upstream. We did some very detailed measurements, and that turned out to be pretty interesting.

Nebeker:

I assume you were doing the numerical models along with this experimental work and validating your models.

Physiology, animal subjects

Mates:

Right. At that point I got more by taking my interest in experimental work and the modeling I had done in the aerodynamics area and sort of merging those together. That work evolved over a number of years. I took another sabbatical around 1980 to learn how to do the animal part of it because at that point I was dependent on collaborators to do any animal experiments, and some of the collaborators I worked with would leave and I would have to get re-established. So I said, “Well, I can do the physiology.”

I spent a year in the physiology department and I learned how to operate on dogs, and do surgery, implant flow meters around coronary arteries, and all those kinds of things. I became much more independent then. Using that background I was able to do a lot more detailed studies of coronary flow distributions. We would take dogs with flow meters on the coronary arteries and insert catheters to measure pressure, and we got much more detailed information on the way in which the coronary flow distributed, particularly in disease states.

I collaborated with a number of students on that, including one student who is still at the University of Buffalo, John Canty, who was a biomedical engineer by undergraduate training, went to medical school, and became a cardiologist. He and I collaborated for a long time on combined modeling and experimental studies of coronary blood flow.

Nebeker:

So, the experimental work would be with dogs, typically?

Mates:

Largely with dogs, yes. We did a little bit of work with goats and sheep. The problem with dogs was that the antivivisectionists were very much opposed to using dogs, so it was very difficult and very expensive to get dogs. We tried other models, but the fact is that the dog is the best model. The only other model that we had any success with were pigs, and for some studies we would use them. The problem with pigs is that they are too big, and if you wanted to do any chronic studies, in which John was particularly interested, where you wanted to instrument the animal and study it over a period of time, you either had to start with a huge pig or take this little pig and grow it into a large one. So the dog was the model of choice.

Clinical applications

Nebeker:

To what extent was this research straight physiology, at least as an objective, where you’re just wanting to understand how the heart functions, and to what extent was it aimed toward understanding diseases?

Mates: 

Understanding the disease was always the ultimate objective, but of course you need to establish an understanding of normal physiology. One of the things that we did in the dogs was to try to predict blood flow with various degrees of coronary occlusion as the artery becomes more constricted.

One of the practical problems with coronary artery disease is that most patients are asymptomatic until the artery is just about totally closed because the circulation compensates. As your artery starts to constrict, the downstream circulation starts to relax to reduce the resistance. So you can occlude, in most cases, over ninety percent of the area of an artery without reducing flow. All of a sudden though, you get to that critical point and then flow starts dropping off.

Nebeker:

Were you able to model that?

Mates:

Yes. We were able to do a lot of modeling on that. The other problem, of course, if you’ve got an artery that is that much occluded and then you have a small clot that sticks, that’s what causes a lot of the acute heart attacks. So, we certainly weren’t all the way there, but I think we made some major contributions to better understanding the disease.

Just about at the end of my research career, I started trying to do some clinical work, where we would actually take measurements in the cardiocatheterization lab and see if we could use the model. That became very difficult because I really needed a good collaborator who was doing catheterizations a lot, and that was very tough to find. We did a few studies, but at that point things were sort of winding down, and that’s where we left it. But it’s still going on, and John Canty is still carrying out a lot of the work that we did together.

Instrumentation

Nebeker:

How about the instrumentation side of this whole research program over the years?

Mates:

A lot of the instruments that we were using towards the end didn’t even exist early on. The sort of instrumentation that we made use of included a lot of ultrasound for making dynamic measurements of length changes. For example, we could sew two crystals on the ventricle and we could watch individual fibers contract. We also used Doppler ultrasound for flow measurements.

Nebeker:

When were you doing that? I think of that as being fairly recent?

Mates:

I would say from the early or mid ‘80s. We used conventional pressure instrumentation and transducers.

Nebeker:

Were there suitable transducers?

Mates:

Yes. They got better as time went on, and actually, at the very end, people were starting to put little piezoelectric crystals on the end of catheters, so you could actually stick them right in the artery. The big problem with the early catheters (in fact, on of the big problems in clinical cardiology) was that they wanted to measure the pressure on the heart, so they’d stick a catheter up through the groin into the heart. You’ve got this fluid filled column which oscillates back and forth. One humorous thing was that when I went in the cath lab for the first time, they were measuring pressure in the ventricles and getting these oscillating traces, and they were recording all the peaks and valleys of these oscillations and trying to figure out what they meant. Well, what they were doing was measuring the natural frequency of the catheter. I said, “I bet that’s what the problem is.”

Nebeker:

The catheter is actually moving?

Mates:

No. There’s fluid in there, and the big problem was that they would take this catheter and fill it full of saline that had been sitting in the cath lab at room temperature. When they would bring up the body temperature, little micro air bubbles would come out of solution. What you had was a mixture of liquid and micro air bubbles that became very elastic, and you could oscillate things. One of the little studies we did was to prove that, in fact, that’s exactly what they were measuring. We can measure the natural frequency of catheters.

Then somebody had the idea to use de-gassed saline, where they just kept it in a vacuum bag. So if you flushed the catheter with that beforehand, all of these oscillations went away. Well, they were pretty impressed, after having collected all of this data for years. So we did quite a number of small studies that I think were helpful to the cardiologists in terms of improving their measurements.

Getting back to your question on instrumentation, the electromagnetic flow meter is basically a coil that you place around an artery and you apply a magnetic field. Since blood is a conductor, you generate an electrical voltage across, and that has a good frequency response, so you can actually measure flow wave forms just with this little thing clipped around the vessel. That was a standard that we used.

Nebeker:

Do you remember when that came out?

Mates:

That had been around for quite a while, before I started working. But ultrasound improved, as did partial instrumentation. Those were probably the main developments.

Computers in experimentation and modeling

Nebeker:

I’m sure also that improved computers made a difference, too.

Mates:

Yes. Three or four years before we finished, we bought a package system that had all the software, so you could just hook up all of your instruments on the screen where all the traces were. Of course, you could store all of the data on the hard drive. When I started working, the first computers were punch card, then we went to magnetic tape, and finally it was all on a disk. Computers were a big asset, not only on the experimental side, but also on the modeling side, because we did a lot of modeling using various optimization methods and fitting parameters to models that would not have been possible by hand, certainly. Even with the early computers, it just would not have been feasible to do that.

Nebeker:

Also, I would imagine that the very mathematical structure of the model was constrained earlier by computing parameters.

Mates:

Yes. You had to have a very simple model, and some of the models that we looked at later were models with ten or fifteen adjustable parameters. You could actually use optimization methods to simultaneously select the best values of those parameters to fit your experimental data. That would not have been practical with the early computers.

Intersections of modeling and experimentation

Nebeker:

So throughout this period, were you working with both the modeling and the experimental and keeping them in touch, so to speak?

Mates:

Yes. There are a lot of people in biomedical engineering that can develop these incredible models, such as electrical engineers who love those models with hundreds and hundreds of resistors and capacitors. Very complex models can simulate anything you want, but the problem is how to select the parameters. My idea was that unless you had a simple enough model so that you could identify the parameters in a particular animal or in a particular patient, the model wasn’t terribly useful from a practical point of view. Although we never quite made it, I thought that eventually we would come to the point where you could take measurements in the catherization lab, model the behavior of the coronary circulation and say, “Okay, this patient has an eighty percent occlusion in this artery,” just from the wave forms. Now we weren’t there, but eventually somebody will be able to do that, for sure.

Nebeker:

I’m sure that a sufficiently complicated model is like Fourier synthesis where you can adjust parameters.

Mates:

I think early on, many biomedical engineers, particularly the ones who were theoretically inclined, loved to model, and they loved these complicated models. The physiologists would look at them and say, “What a bunch of garbage. This is of no use to me at all.” There was a certain prejudice. I noticed that frequently with the NIH study sections. People would see a model and they would say, “Well, that’s clearly not worth funding because it’s just more nonsense.” The physiologists were very distrustful early on of the modelers. That’s changed over a period of time because I think they realized that there was a lot to be gained with looking at models.

Nebeker:

Also, I imagine that from some sort of ad-hoc modeling to modeling that corresponds in its mathematical description to physiology. Is that the case?

Mates:

What we were doing was taking a much more macroscopic modeling approach. In other words, in principle you could say, “I could model an individual sarcomere. I know how that works. Then I can put that together with a bunch of other sarcomeres into a muscle fiber. Then I can look at this complicated structure of the ventricle and I can put all of that together and predict what will happen.”

Even today, that’s not a practical approach because there are literally millions of these sarcomeres arranged in all sorts of different geometries in the heart, and so you can’t really build up from the bottom up. What you have to do is make some compromises and say, “Let’s assume a certain idealized geometry, and let’s assume that the average sarcomere is pointed in this direction, and we know how it works.” You can build up in that sense, but to actually model every sarcomere in the heart is not practical because we don’t even know in a particular heart what the microstructure really is. We only do in a general sense.

Nebeker:

But isn’t having adequate description at each physical level the long term objective? Like you have the sarcomere modeled, then you have the muscle fiber modeled, and you’d be able to at least show that one can derive the next higher level from it.

Mates:

Sure. But it’s going to be, even in the end, a very idealized picture. Certainly having the sarcomere mechanics, knowing how the sarcomere is contracted and what influenced the force/velocity linked relationships was a major breakthrough in terms of modeling the ventricle because without that information, you didn’t really know where to start.

Nebeker:

I can also imagine that if you’re modeling at a higher level, the cardiologist is skeptical of the models. I suppose that’s where the experimental work comes in. You’re showing that you vary all of these parameters and the model still works.

Mates:

Yes. Ideally, to apply this to the clinical arena, what you would be able to do is take a particular patient and use the model to describe the behavior of that patient’s heart. For example, then you could study the effect of different drugs or surgery on improving function and you would have a more quantitative way.

Typically, even today, the gross measure of the efficiency of the heart is something called the ejection fraction. You measure the volume of the ventricle when it’s full, measure the volume when it’s empty, and you calculate how much of the blood was ejected. That’s a crude measure, but it’s still very effective. In other words, if you find a patient who has an ejection fraction of twenty percent, that heart is pretty sick. In a healthy patient, it’s more like two thirds or three quarters. You’d like to do better than that.

Nebeker:

So, you’d like to be able to take a few measurements on this patient and have a better understanding of what’s going on.

Mates:

Right. That’s sort of where we were when I dropped out of the picture. We were beginning to be able to do that.

Nebeker:

How much of your work was dealing with cardiac muscle, and how much dealt with the flow behavior in the coronary arteries?

Mates:

I would say the majority dealt with the flow behavior, but we did quite a bit of work with muscle, too. It would be like two thirds and one third.

Nebeker:

They seem like different areas to me.

Mates:

They’re both mechanics, and if you understand basic fluid and solid mechanics, they are both applications of that. Take Bert Fung for example. He’s basically done work in all areas of biomechanics. He’s developed models of totally different things. His work is far more wide ranging than anything I did, but it’s an example of how you can use mechanical ideas.

Nebeker:

So once you’re able to model and experiment with one, it’s not…

Mates:

No, it wasn’t that big of a switch.

Cardiac muscle and coronary circulation findings

Nebeker:

You’ve named a few things, but just to be sure we get it on record, what do you regard as the most important findings with respect to the cardiac muscle in your research?

Mates:

It’s hard to pick out one thing. We didn’t have any Nobel Prize winning breakthroughs. I think we were able to suggest measurements that could be made better, I mean, simple things like improving the performance of catheters in the catheterization lab. That’s a very simple thing, but that saved them an enormous amount of time in analyzing their data and made them understand better what they were measuring. That was just a very simple idea.

With some of the animal models we were able to clarify the behavior of coronary artery disease in the sense of why it was possible to occlude an artery ninety or ninety-two percent without affecting blood flow. We certainly made some contributions to that, not that we were the only ones, but I think that was an area that was important to the clinical cardiologists.

Nebeker:

Were instrumentation techniques that you developed adopted by others?

Mates:

Yes. I think some of them were.

Audio File

MP3 Audio
(278 - mates - clip 2.mp3)

They were mainly small improvements in a variety of areas, but certainly some of them were. One of the things that we developed that is being used by other people now (in fact, we used to get requests for it) was a control system that you can run blood through that could produce the kinds of pressure and flow wave forms that were seen in the coronary circulation. We initially developed that for use in this model coronary artery that I mentioned, then we scaled it down and we actually were using it to profuse isolated hearts from dogs with different kinds of pressure and flow profiles. That way, we could sort of uncouple the behavior of the ventricle itself from the coronary artery.

One of the things that complicates the coronary circulation is the fact that the heart is beating. So every time the heart contracts, the pressure goes up. But simultaneously, those vessels downstream get squeezed by the heart muscle. They’re going through the heart. So it’s not like a simple hydraulic system, and that makes it much harder to understand the phasic relationships between pressure and flow because the resistance is varying as the pressure varies. We were able to uncouple those with this device we had because we could stop the heart from beating for a moment, and then still parfuse it with the same pressure and look at the difference in flow patterns.

So we were able to get some information about the relative importance of the squeezing effect and the pressure. That device, called the hydraulic serovalve was a feedback control system where you would prescribe a wave form, and then it would follow electrically whatever wave form that you prescribed using electrical engineering principles. We could simulate all kinds of wave forms. We actually did a frequency response in the coronary circulation by applying sinusoidal input pressure of varying frequencies and measuring the amplitude of the flow response, and we could determine the frequency response of the circulation under different conditions, beating and non-beating. So that’s basic mechanics, but also with the electrical technology to make it work.

Nebeker:

Something that I’ve encountered in quite a few cases is where people like you have developed some device that they thought of as for their own experimentation, and then that auxiliary device itself becomes something for the world.

Mates:

I wouldn’t say that it’s used all over the world, but we had several people who have said, “Would you send me a set of drawings because we want to build one of these things?”

Nebeker:

This is sort of like a standard signal generator.

Mates:

Yes. It’s a hydraulic signal generator. We did it because we were looking for ways to better understand the dynamics.

Graduate students and collaborators

Nebeker:

What people did you work with? Of course, you had graduate students.

Mates:

I had graduate students from engineering. I worked with a lot of cardiology residents who came through our lab. I had several collaborators. I mentioned a couple of them. Over the years, young cardiologists would come and train in the laboratory. I worked with a number of them.

My most important collaborator, I would say in the whole bio area, were David Greene, who got me started; Fran Klocke, he and I worked together for over twenty years until he got lured away by Northwestern University a few years ago; and John Canty, who started out as an undergraduate biomedical engineer in our lab and eventually became chief cardiologist. They were probably the three most important collaborators.

Center of Biomedical Engineering; biomedical engineering as a field

Nebeker:

At some point you became director of…

Mates:

We had a Center for Biomedical Engineering at the university, which was kind of low key. What we basically did was try to make people aware of what was going on in the university. We did a lot of work getting local companies involved with individual researchers and particular projects that they were interested in. We made people in the western New York community aware of what was going on at the university.

We also would sponsor seminar series and that sort of thing, but we were very small and we didn’t have much funding compared to other centers at the university. I think it made it easier for students to find faculty who had similar interests. We had a student chapter of the Biomedical Engineering Society, and I was the faculty advisor for that. Students would come to me with particular interests and I would point them to individual departments.

I don’t know if you’ve encountered this with other people as well, but one of the problems with biomedical engineering is that it’s a very strange field in the sense that on the one hand it’s specialized, but on the other hand it’s very broad. Almost every one of the engineering sciences has biological applications. So one of the things that I should have thought of when I started that center and that became clear over a period of time was that there were not a whole lot of common interests. We had faculty in EE who were interested in signal transduction and electrical propagation of the nervous system. Our dean of engineering, George Lee, was interested in the elastic structure of the lung.

Nebeker:

He was a trans-civil engineer.

Mates:

Yes. He was in engineering mechanics. So we had, even within mechanical engineering, a range of interests in solid mechanics, fluid mechanics, and control systems. You can’t train anybody to do all of that stuff, and I think that’s the problem that has haunted biomedical engineering education for years and years. I know that IEEE has been, and I don’t know if they still are, the sort of lead society in setting up the accreditation standards for biomedical engineering programs. That’s a huge problem because if you look at various biomedical engineering programs around the country, they have very little in common with each other.

Nebeker:

That’s right. I know that the UC San Diego program was more in mechanical, and so it looks very different from places where the emphasis is on the electrical.

Mates:

University of Pennsylvania, for example, was basically started by EEs. I think Rutgers was EE-based.

Nebeker:

Yes. There are some very important EE biomedical engineers there. And at SUNY Buffalo, was it more mechanical there?

Mates:

I would say it was divided about equally between mechanical and electrical. There were some people in industrial engineering. Ergonomics is really biomedical engineering—workplace design. We had a couple of people there who would consider themselves biomedical engineers. We had a couple in civil engineering, also. In chemical engineering, not so much genetic engineering but the whole area of molecular and cellular level studies have a lot of chemical applications. We had a couple of people in chemical engineering who were studying cells, for example.

Nebeker:

Did you set up the center?

Mates:

George Lee, who is the Dean of Engineering asked me to do it, and I was the director of it for four or five years before I retired.

Nebeker:

So you tried to learn about all the people on campus?

Mates:

Yes. I tried to know what was going on. We would publish an annual report listing grants and publications. So I sort of knew where all the activity was. I didn’t understand all of it, but I was able to match students with faculty and I would occasionally get a call from somebody in local industry. We had a number of biomedical device companies locally. They would say, “We’re interested in this area. Is there anybody at the university we can work with?”

Nebeker:

Was there an undergraduate biomedical engineering program?

Mates:

No. And I strongly resisted that, and I would continue to resist it today because I think it’s very hard to establish a program that is representative of what’s going on. It’s almost impossible, from my point of view, to train somebody in four years to be a practicing biomedical engineer. The graduate level is fine. But one of the worries I have about biomedical engineering is that the Whittaker Foundation, which has been a great friend of biomedical engineering, is creating a field because they have all this money. I don’t know how many Ph.D. programs have been established using Whittaker funds but a lot. They’ve supported the Biomedical Engineering Society; they gave them a million dollars. The American Institute of Medical and Biological Engineering is a Whittaker funded operation. The Whitaker Foundation will go away in a few years, and what’s going to happen? I don’t know.

Nebeker:

The idea is that this field is right for development and it’s seed money. But you have some doubts?

Mates:

I have some doubts about whether biomedical engineering is going to be a longstanding activity, at least at the undergraduate level. If you go back in history and look, for example, at the Engineering Education Journal over the years, there are many fields. I don’t have a complete set, but I’ve gathered some data. In Buffalo we lived through programs in engineering science and nuclear engineering. We probably started those too late and suffered with them until we finally took the step of eliminating them, which is a very painful thing to do. I just have the feeling that we may be overbuilding biomedical engineering at the undergraduate level.

Nebeker:

At the undergraduate level, it doesn’t make sense to train people for this vast field in terms of the technology.

Mates:

I don’t think you can do it. It wasn’t clear to me that there was much of a demand from industry for biomedical Bachelor’s degreed people, because if you’re making catheters for example, you need materials specialists, people who know how to manufacture things. If you’re making prosthetic joints, again you need material specialists and manufacturing people. There’s some biology involved, but largely at the research level. When it comes down to actually fabricating a piece of equipment, you really need the traditional kinds of mechanical, electrical and material science people.

Nebeker:

I’ve wondered about that myself.

Mates:

You’re going to find people who will violently disagree with me.

Engineering science; practical education

Nebeker:

I hadn’t heard the comparison with engineering science. Tell me a little about that field.

Mates:

Engineering science was a program that was a response to Sputnik, basically. Everybody said, “Engineers don’t know enough science. They’re too much nuts and bolts oriented.” I can’t even tell you who started the field, but it was in the early ‘50s that it got started as an undergraduate program, and there were quite a number of those programs around the country.

Nebeker:

So the idea was that a person could prepare for general engineering.

Mates:

Audio File

MP3 Audio
(278 - mates - clip 3.mp3)

Yes. And some of them were actually called General Engineering, some were called Engineering Science, and some were called Engineering Physics. They were all designed to have more mathematics and more science in the engineering program at the expense of the practical applications. Industry never understood what they were, or had ever heard of them. So, what happened was that the students who went into those programs and who went on and got PhDs were fine. I think at the Ph.D. level, the field of specialization is not really a critical issue because you will change over the course of your career. You will learn many more things than you ever learned in graduate school over a 30 year career. Look at my career. I went from turbo machinery to hypersonic aerodynamics to blood flow. It didn’t really matter what my Ph.D. was in.

But at the undergraduate level, a lot of people have to go out and find jobs. If you take a program like UC San Diego or Johns Hopkins, almost all of those students go to graduate school or medical school. So an undergraduate degree in biomedical engineering is probably very good training for medical school. My daughter actually went that route. She got a degree at Michigan State in systems engineering, and she went to medical school and that was good training.

On the other hand, to go out and expect to find a job in four years is difficult. There are a lot of schools that have used biomedical engineering as an undergraduate program to attract students because it’s very popular and very trendy. There’s a school that I won’t name because I don’t want to point fingers, but their biomedical engineering is the largest undergraduate program in the engineering school. And it’s not a school that attracts all Ph.D. caliber students. Now again, people will certainly disagree with me. I don’t see this as a long-range established field, probably because it’s so broad. There are many examples. Even mining engineering, textile engineering, and automotive engineering are all programs that have sort of petered out over the years because there just wasn’t a big enough demand at the undergraduate level.

The trouble with any program is that if you start turning out Ph.D.s, they want to teach and they want to have undergraduate programs because that’s what provides the bread and butter. So you will see over a period of time that all these Ph.D. programs will spawn undergraduate programs. I have some real concerns about that. I’ve had big arguments with a lot of people, and I’m probably in the minority. I won’t be around long enough to see who was right.

Nebeker:

I’m very grateful for your opinions on this.

Mates:

I think Whittaker has done a great job with biomedical engineering. I just worry that they’ve overbuilt it. That’s all.

IEEE, ASME

Nebeker:

At any level, it certainly impresses me how diverse the field is, and I wonder if you can comment on that. I know best the IEEE EMBS. What is the structure in ASME?

Mates:

ASME has divisions. They’re not a separate society like IEEE, but there are divisions within the society. Back in the ‘40s, they established something called the human factors division, which was more sort of industrial engineering.

Nebeker:

Out of World War II, that was a big interest.

Mates:

Right. I think that was changed to the bio-engineering division in the early ‘70s. The name was changed. That was just about the time that I was getting active in the field. It has become one of the fastest growing divisions within ASME in terms of membership.

The Journal of Bio-mechanical Engineering has grown enormously. In fact, I was the Vice President of the Board of Communications for ASME for three years, and during that period, which ended two years ago, we took The Journal of Bio-mechanical Engineering from a quarterly journal to a bi-monthly journal because it was growing so fast that we just couldn’t publish all of the papers. That was the first journal in the ASME to get published six times a year. It’s been very successful. Bert Fung was the founding editor, and the programming at the major winter meeting, which is now the Congress, has become one of the biggest of the divisions in terms of sponsoring sessions. They were very viable in the summer meeting. But I would say the emphasis there has always been mechanics and materials—solid mechanics, fluid mechanics, orthopedic biomechanics, a lot of that, prosthetic devices. We’ve never branched much into the typical IEEE area. The EMBS is all over the map. They do mechanics and the traditional electrical stuff, but they also do practically everything else.

Nebeker:

You mean the EMBS?

Mates:

IEEE.

Nebeker:

They claim to be very broad and they certainly have people outside the typical IEEE technologies, but I think still their focus is on the electrical.

Mates:

Their focus is very broad. Now, the Biomedical Engineering Society is a very small but very active group. I would say they have tended to be more mechanics-oriented than anything else. They are probably closer to ASME than they are to IEEE.

Nebeker:

And how about the connections with the chemical? That’s also a huge branch of biomedical engineering.

Mates:

Yes. I don’t know much about that. I’ve never gotten involved too much with that. I know we did have quite a bit of activity at Buffalo, and certainly, that’s probably the fastest growing activity now within bioengineering with the cellular and molecular emphasis, but I don’t know a whole lot about it. I know that chemical engineers have a section of biomedical engineering.

Nebeker:

Yes. It’s very large. But my own feeling is that the IEEE people had more to do with mechanical people than with the chemical people.

Mates:

That’s probably true, and IEEE was probably the first. They got started around ‘52, and we didn’t get started in the ASME until the ‘70s. I know that IEEE, from the very beginning, was the so-called lead society in establishing accreditation guidelines for biomedical training programs. The Biomedical Engineering Society has been wildly jealous of them for years. In fact, I was on the BMES board of directors for a while, and every year, they would say, “Wow. We’re going to take over the accreditation,” and somebody would say, “Well, it’s going to cost you hundreds of thousands of dollars a year to do that.” They just didn’t have the resources to do it. They may now that Whittaker has given them a lot of money. I don’t know.

Disciplinary definitions and funding

Nebeker:

I’m always interested in discipline formation and how one actually functions. Is one out there as a mechanical engineer feeling principal affiliation to the ASME and dealing most with mechanical engineers, or say that you get into the heart field and associate with the cardiologists?

Mates:

That’s another problem with being a biomedical engineer. You have to have funding, and that comes from the medical side, from NIH primarily. In order to get grants from NIH, you have to be known in the medical community, publishing in the medical journals. The Journal of Bio-mechanical Engineering is a respected journal, but it doesn’t get you funding. So you’ve got to get published in their journal and you’ve got to go to their meetings. I would normally go to the Physiology Society meetings and to the Heart Association meetings, at least, and sometimes the American College of Cardiology, in addition to the ASME. And a lot of my publications were in medical journals because that’s where people would read them. Some of the work that I did in modeling of the coronary circulation, which was totally done by engineers, was published in the American Journal of Physiology, because that was where you had to get them published in order for people to know about you. They were the ones who reviewed your grants. There were no active collaborators on the medical side.

Nebeker:

Most of the people that I talked to were trained as engineers, but some of them are trained as MDs. Jim Bassingthwaighte comes to mind. I was just up in Seattle talking to him and a couple of others.

Mates:

He went through an MD/Ph.D. program. He had the biophysics training. I think his Ph.D. was in biophysics.

Nebeker:

But then, getting into what’s now biomedical engineering, one can come from the science and physiology side and doing work that’s regarded as biomedical engineering, or one can come on the engineering side.

Mates:

Yes. And there’s tremendous overlap between biophysics and biomedical engineering. There’s probably a fundamental difference between people who are trained as scientists and people who are trained as engineers. People who were trained as scientists had this classical hypothesis-driven research basis, and you can’t really get a grant, I discovered early on from NIH, unless you have a hypothesis. An engineer will say, “Gee. That’s interesting. Let’s study that and see what happens.”

The person I know that feels the most strongly about that is Geart Schnoidt at UCSD, who is an engineer. He said, “This is ridiculous.” He said that scientists constrain themselves far too much because they have to have hypotheses to test, and he said that they’re missing a whole lot. I agree with him on that. I think with a lot of things you think, “Well, yeah. This is really curious. Let’s just study it and see what happens.” You can’t get funded by NIH that way.

So it’s a different approach, and I think a lot of the people who are successful in the field are acting more like scientists than engineers. The greatest fun I had was building gadgets and developing experimental procedures, but you don’t really get money to do that kind of stuff. That’s something you sort of have to bootleg on the side.

Nebeker:

Or it might get published in the Instrumentation Society’s journal.

Mates:

That’s right.

Comparison of biomedical engineering and physiology

Nebeker:

You’re a good person to ask about this. It seems to me that a lot of work done by people who are today called biomedical engineers, and a lot of your work is in this category, looks like it’s at least in the agenda of classical physiology—trying to understand blood flows, cardiac anatomy, and so on. And what traditionally made it biomedical engineering is the engineering approach to that, either developing instrumentation. Earlier it was more unusual, as with numerical modeling and computer modeling. Often it’s called “the systems approach,” regarding it as a system, whereas the traditional physiologists are more for going in at a particular level and trying to understand that.

Mates:

There certainly is a large overlap. There are people in physiology departments, such as a physiologist we had in Buffalo who had a Ph.D. in biomedical engineering, who was teaching and working in a physiology department. Physiology has become a lot more mathematically and…

Nebeker:

Instrumentally sophisticated?

Mates:

Well, I was thinking more of modeling. They’re doing a lot more sophisticated modeling of physiological systems, and some of that work is indistinguishable from work that biomedical engineers are doing. Some people that I knew at Buffalo who were trained as physiologists had acquired the mathematical background over the years and the ability to do that kind of modeling. It was very sophisticated stuff. That’s probably inevitable over a period of time, that when one field sees something from outside that’s useful to them, they grab on to it.

The other thing that has happened in physiology, which is even more dramatic, is the shift in emphasis from the sort of systems physiology, the large scale stuff, down to cellular and molecular. A lot of physiologists worry that there’s so little systems physiology research being done, probably because they can’t get funding for it any more, who’s going to teach the medical students how the heart works? All the lectures will be about the cellular and molecular physiology of the heart, but nobody will actually ever study the intact organ. That’s a real concern in medical education.

But the payoff has been so great in some of these studies that you can really get a much more fundamental understanding of how things work when you go down to the smaller level. But you give up something, too. You give up the big picture.

Nebeker:

Do you think that there is a convergence through time in the classical physiology stream and the biomedical engineering in that area?

Mates:

Certainly in the areas where there’s overlap. But it’s more now with the chemical engineering part rather than the mechanical engineering because the emphasis is now on the small stuff, the cells and the molecules. That’s where the chemical engineers are much better equipped than the mechanical engineers to deal with it.

Nebeker:

But when one looks at articles on the heart or on coronary circulation, can you tell at a glance that this is a physiologist and this is a biomedical engineer?

Mates:

Not anymore. You could have 30 years ago, but not anymore, no. I would say that some of the work is so close that you would never know the difference.

Neuro-muscular disease rehabilitation

Nebeker:

Another area in which you’ve worked is rehabilitation of patients with neuro-muscular diseases.

Mates:

Yes. That was a very small part of what I did and very late in my career. I probably wouldn’t have even thought to mention it. I had a friend who was interested in rehabilitation, and we developed some devices for training patients who were either post-stroke patients or they had other neuro-muscular weaknesses. It was basically an instrumented weight bench with which that you could make measurements and you could apply increasing degrees of load. It also had patients’ feedback. That was one of the unique features of it, and that was not my idea.

Nebeker:

You very carefully controlled the rehabilitation.

Mates:

Yes. We were going to build a device that could it to the patient who could take it home after a stroke. The patient would get on this thing in the morning and say, “Okay. Today, we’re going to do this.” Then it actually had voice feedback to tell the patient to push a little harder and things like that. The idea was to be able to measure the muscle strength of the patient over a period of time, and it had a feature that recorded the data each day, so that when the nurse practitioner, the physical therapist, or whoever was looking after the patient would come and get a history of exactly what happened with the patient. I had a couple of students who worked on that, and that got us into some other devices.

For example, we built an instrumented set of stairs for patients with gait problems where we had force plates on the stairs and they could climb up the stairs and we would measure the distribution of force that they had with the objective of understanding their gait, but also looking at improvement after some kind of surgery, for example.

Nebeker:

So you had an accurate record of how they walked up the stairs before and after.

Mates:

Right. And we could adjust the pitch of the stairs. These were mainly senior projects that the students did as undergraduates in a mechanical design course. We did a number of little things like that. I had sort of forgotten about that. I always considered that a sort of side issue in my work, not the main thing because we never had much funding for that. It was fun.

Social value of biomedical engineering

Nebeker:

Are there areas that I haven’t asked about that you’d like to comment on? Areas of your own research?

Mates:

I don’t think so. I think you’ve pretty well covered the ground. But to sum it all up, it was certainly some of the most rewarding work that I ever did. I think that’s the thing that makes biomedical engineering attractive to students. In other words, it’s much nicer than building missiles. It’s something for humanity.

Nebeker:

Or even bridges. It would seem so cold.

Mates:

Or automobiles that are going to pollute the atmosphere. It’s a very attractive field. That’s a worry because I think there are unscrupulous academics around who will do anything to attract students, and they have unrealistic expectations of what’s in it at the end of the road. So I have concerns about it. There is no question in my mind, though, that electrical engineers, mechanical engineers, and chemical engineers will be increasingly involved with biomedical problems as time goes on. There’s no doubt about that at all.

Biomedical engineering as a field

Nebeker:

And what about the coherence of biomedical engineering as a field?

Mates:

I worry about it.

Nebeker:

Is it there?

Mates:

I don’t think so. I have doubts about whether it will ever be there because of the breadth. If I were to take a guess and look 50 years down the road, my guess would be that biomedical engineering as a separate discipline will be very small, but within IEEE, ASME, and chemical engineering there will be increasing amounts of activity. That’s what I think is going to happen.

Nebeker:

Thank you for the interview.