Oral-History:David Geselowitz
About David Geselowitz
David Geselowitz was born in Philadelphia and attended the Moore School of Engineering at the University of Pennsylvania. Upon graduation, he accepted a position as assistant instructor at the Moore School as he went for his Ph.D. He then got involved in research with electrical noise, funded by the Signal Corps. He worked on a project with Ed Hawthorne involving metal conductors on windshields and wrote his master’s thesis on the shielding of glass by a metallic film manufactured in Pittsburgh. He took a position replacing Robert H. Okada in Herman Schwan’s electromedical lab at the Moore School, though Okata remained an advisor of his. His dissertation concerning the potential field on the body in terms of cardiac sources became a classic in Dr. Eugene Garfield’s Science Citation Index. He obtained his Ph.D. in 1958. Soon after, he was offered a job at Bell Labs but continued working for the Moore School due to an interest in teaching and the academic world in general, despite what he recognized as an opportunity for relatively unrestricted research by industry standards at Bell. He worked with Paul Langer for a number of years on signal domain and Stan Brillerand and was guided by S.A. Shelkunov in work on potential and multipolar expansion.
Geselowitz became the senior person under Herman Schwan, through which he got involved with the IRE, a predecessor to the IEEE, which he believes was a key organization in advancing biomedical engineering, along with other groups such as the ASME . He and Okada, as well as others, continued the work of four major labs that made the core of what he refers to as “the first generation” of biomedical engineering. This work eventually led to achieving results from the cellular activity in the heart.
He also edited Transactions for a time and was a key factor in establishing a Ph.D. program at Penn State for biomedical engineering in 1971. He and a student, Walter Thomas Miller came up with the Miller-Geselowitz model, accounting for a normal electrocardiogram. He retired in 1998 with the Ph.D. program he helped establish flourishing, and many schools now offer biomedical engineering to undergrads, although Geselowitz thinks it should be relegated to graduate work. He continues to attend seminars and faculty meetings. He has no regrets with his choice in career, as he finds biomedical engineering fascinating.
About the Interview
DAVID GESELOWITZ: An Interview Conducted by Michael N. Geselowitz, IEEE History Center, 31 August 2004
Interview #449 for the IEEE History Center, The Institute of Electrical and Electronics Engineers, 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, 39 Union Street, New Brunswick, NJ 08901-8538 USA. 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:
David Geselowitz, an oral history conducted in 2004 by Michael Geselowitz, IEEE History Center, New Brunswick, NJ, USA.
Interview
Interview: David Geselowitz
Interviewer: Mike Geselowitz
Date: 31 August 2004
Place: State College, Pennsylvania
Childhood, educational background
M. Geselowitz:
This is Michael Geselowitz. I am in State College, Pennsylvania, interviewing Dr. David Geselowitz on August 31, 2004. You are a Life Fellow of the IEEE and a Fellow of IEEE's Engineering and Medicine Biology Society. Tell me about your childhood and how you got interested and involved in engineering.
D. Geselowitz:
I was born in Philadelphia and grew up in a neighborhood called Strawberry Mansion. I was always interested in math. At one point someone gave me a book by Lancelot Hogben, Mathematics for the Million, that I read from cover to cover. The book presented a number of topics which were not to be covered in mathematics in school for several years. I was intrigued, but when it came time to make a choice of a career, after talking to people, I felt that there were not many opportunities in mathematics. Engineering seemed to offer more opportunities, and I was told that electrical engineering probably involved more mathematics than other branches of engineering.
M. Geselowitz:
What year did you graduate high school?
D. Geselowitz:
I graduated from Central High School in Philadelphia in 1947. Central High School was marvelous. It was a citywide academic high school.
M. Geselowitz:
Did you also have good academic advising?
D. Geselowitz:
Yes.
M. Geselowitz:
Do you think that in average high schools in middle America advisors would have recommended electrical engineering to students that were good in math and science?
D. Geselowitz:
I do not know for sure. I should note that an uncle of mine was an electrical engineer.
M. Geselowitz:
Okay, so you already knew about electrical engineering.
D. Geselowitz:
Yes. The high school I attended was excellent. As a matter of fact most of the teachers at that school had Ph.D.s. This was the end of the Depression and a lot of people who had doctorates couldn't get other jobs and they took jobs as teachers.
Electrical engineering studies at the University of Pennsylvania
M. Geselowitz:
Where did you go to college to pursue electrical engineering?
D. Geselowitz:
I went to the University of Pennsylvania. Electrical engineering at Penn was in The Moore School of Electrical Engineering. It was a strange arrangement. It was a separate entity.. I'm not entirely clear about it, but I believe [Alfred Fitler] Moore bequeathed money to establish a school of electrical engineering and the Trustees decided the most effective way of implementing this was to affiliate with the University of Pennsylvania. However the school maintained a separate Board of Trustees and was headed by a person with the title of dean, so that The Moore School, while it was an integral part of the University Pennsylvania, maintained a lot of its own identity.
M. Geselowitz:
Even though it had an undergraduate program it was like a medical school or law school where there is a separate dean and separate board of trustees?
D. Geselowitz:
I don't know that the medical and law schools have separate boards of trustees.
M. Geselowitz:
They do at Harvard.
D. Geselowitz:
Not at Penn as far as I know. I don't think those other schools had separate Trustees.
M. Geselowitz:
Did The Moore School of Engineering have a separate admissions program?
D. Geselowitz:
Not of which I am aware. Everything is through the University of Pennsylvania except they had this particular arrangement. There were departments of chemical engineering and mechanical engineering, and because of this the University of Pennsylvania had a school of electrical engineering and then a Towne School for all the other engineering.
M. Geselowitz:
That's not very symmetrical.
D. Geselowitz:
Right.
M. Geselowitz:
What did you study there?
D. Geselowitz:
I was a student in electrical engineering. The curriculum was, as I guess has always been the case in engineering, pretty well determined. There are a lot of required courses and then there are electives and that's it. One had to take a Humanities elective each semester. The first year or two one took math and physics.
M. Geselowitz:
You were eventually to make your mark in bioelectrical engineering. At that time, as far as I know, that was not even recognized as a field.
D. Geselowitz:
Right.
M. Geselowitz:
When did you become aware of bioengineering and start taking courses in that? How did that direction come about?
D. Geselowitz:
In The Moore School there had been an effort going back a number of years – though I don't know how many – which involved biomedical related research. There was an electromedical laboratory associated with The Moore School, which was headed by Herman Schwan. There were two or three major thrusts to this laboratory. When I was an undergraduate one of the undergraduate instructors, Ernie Frank, was involved in the field of electrocardiography. However, as an undergraduate there was no particular exposure to it. There were no courses involving biomedical aspects. At that point there were no research opportunities on the whole in that undergraduates did not work in laboratories of their professors, though I think that is more common today. As an undergraduate I was aware of this but did not have any opportunity to participate in this effort.
Graduate studies, U. of Penn.
Funding, student culture
M. Geselowitz:
What did you decide to do with your career as you approached graduation?
D. Geselowitz:
I was considering getting a job in industry but I was also considering going on to graduate school. I got and accepted an offer to become an assistant instructor at The Moore School, and that paid my tuition. I do not recall all the aspects of it, but there was a special interdisciplinary program at the university under whose aegis one could go to medical school. One of the professors spoke to me about that possibility, but I opted instead to go for a degree in electrical engineering. As far as I know, very few people went through that interdisciplinary program.
M. Geselowitz:
Were you initially thinking of getting a master's or did you at that point figure you would go on for a Ph.D.?
D. Geselowitz:
My interest was in a Ph.D.
M. Geselowitz:
Why was that?
D. Geselowitz:
I'm not sure. My background was such that I had a general interest in learning. I picked the field of electrical engineering and was interested in the Ph.D.
M. Geselowitz:
Tell me about your graduate program. At that point you would have had more opportunities to do research with the faculty.
D. Geselowitz:
Correct. The Signal Corps funded my initial research. I got assigned to a project. By and large I think that if you didn't have this support you weren't in graduate school. Each graduate was assigned to one of the funded research projects in The Moore School. I got involved with one that had to do with electrical noise and noise in general – signals and noise and signal-to-noise ratio and how noise is measured.
M. Geselowitz:
Who was the professor in charge of that?
D. Geselowitz:
Ralph Showers was the professor in charge of that. I started in 1947 with one of the first classes that had veterans who were studying under the GI Bill, and most of my classmates were veterans. About 25 percent of us were high school graduates and 75 percent were veterans. Those veterans were very serious students. Here's one incident. The freshmen had to wear beanies, so I got a beanie. My first or second day there I was walking on campus with a classmate who was a veteran. A sophomore came up to us and said to him, "Where's your beanie?" I don't recall what the exact words of my classmate were, but the beanies disappeared.
M. Geselowitz:
That was a transitional generation in terms of higher education in the United States.
D. Geselowitz:
Yes. All that silly stuff disappeared for at least a few years.
Master's research
D. Geselowitz:
I was involved in this research with noise, and with a master's degree I got involved in another project with Ed Hawthorne on measurement of the shielding effectiveness of thin sheets of metal. It actually had to do with windshields on aircraft. They put a conducting metal on windshields and put current through it to heat it up. This acted to deice.
M. Geselowitz:
That is done with rear windshields in cars.
D. Geselowitz:
Yes. This was a metallic film that was manufactured by a company in Pittsburgh. It did not optically cut down the light very much. I believe the company was called Pittsburgh Plate Glass [Company]. Someone got the idea that maybe this would be a good electrical shield. The question was, "What are the shielding properties?" I studied this over a range of frequencies up into the microwave to determine the shielding of this glass, and that was my master's thesis.
M. Geselowitz:
Did that end up having an application that was useful to the military or otherwise?
D. Geselowitz:
We reported this to the glass company. I guess they were the ones who were supporting this research. I don't think anything ever came of it. The strange thing is that a number of years later, fairly recently, I got involved with a project in which there was interest in the transmission of energy through the metallic enclosure for the telemetering device with the artificial heart. We used the approach that I had in my thesis.
M. Geselowitz:
Interesting.
D. Geselowitz:
Another thing is, in recalling this I do not remember exactly how I came up with the approach that I took, but I think it was in a book by [James Hopwood] Jeans. At that time in electromagnetic field theory there was the work of [James Clerk] Maxwell, and several people published textbooks using Maxwell's book and updating the notations he used for vector analysis.
M. Geselowitz:
Right. Maxwell's equations in the original form are said to be incomprehensible to all but a few people.
D. Geselowitz:
No, I would not say that. At any rate I did find in Jeans' classic book on electromagnetic theory that there was a result that had been developed by Maxwell for what happens with eddy currents. Basically if you consider a small source and a conducting sheet, when the source is turned on it results in circular currents, eddy currents, in the conductor. There was a derivation for what would happen under this circumstance. I went to Maxwell and found this derivation, which was beautiful. In Maxwell's book I also found a marvelous summary of electricity and electromagnetism and it talked about the experimental basis of the laws. Maxwell's work was excellent, and the result that I somehow stumbled upon and used was a derivation of Maxwell's.
M. Geselowitz:
You went back to the original sources.
D. Geselowitz:
Yes.
M. Geselowitz:
Do you think Maxwell should be used more directly in teaching rather than these later textbooks?
D. Geselowitz:
Probably.
M. Geselowitz:
I can't remember back to my theory too well. I don't use it much.
D. Geselowitz:
There have been more and more books since then.
M. Geselowitz:
I remember that originally people had trouble with Maxwell and I think [George M.] Whitesides who originally came up with a different way of expressing them that was more palatable to many people.
D. Geselowitz:
Yes, Maxwell did not use divergence and curl. He spelled it out, but the equations were there. The curl of J is spelled out: partial with respect to x, y, z.
M. Geselowitz:
Interesting.
D. Geselowitz:
And that was my master's research.
Ph.D. research, electrocardiography
M. Geselowitz:
What did you decide then to do for your Ph.D.?
D. Geselowitz:
One day Bob [Robert H.] Okada, a fellow student, approached me. Bob was working on problems in electrocardiography. To a large extent Herman Schwan's interest was in the impedance of biomatter – the conductivity as a function of frequency through a tremendous frequency range, from very low frequencies up into the microwave. However there had been another effort in the Electromedical Laboratory involving an engineer and a cardiologist. The engineer at the time was Ernie Frank, whose name I mentioned earlier.
M. Geselowitz:
Right.
D. Geselowitz:
Okada had taken over from Ernie Frank. Basically there was one electrical engineer working with a cardiologist. I think Ernie Frank was the third one in the line, unless I'm missing someone.
M. Geselowitz:
And now Bob Okada was the fourth?
D. Geselowitz:
Yes. Okada got his Ph.D., was continuing the research, and was thinking of leaving. He had a product he wanted to commercialize, which I'll come to in a minute. He wanted someone to continue this effort. I would be the fifth one in this chain.
M. Geselowitz:
In this lab most people were working on other projects but there was always just one engineer on this specific project with the electrocardiogram?
D. Geselowitz:
Yes, going back about ten years. Ernie Frank, the predecessor of Bob Okada, made major, major contributions to electrocardiography. The couple of people that preceded him did not make a major impact. I knew Ernie Frank. I had taken a course with him as an undergraduate. I don't recall taking any graduate courses from him. When Bob Okada came to me and told me he was seriously considering leaving and asked me if I would be interested in taking his place I actually probably thought about it for 15 seconds before I decided yes and then thought about it overnight.
M. Geselowitz:
What about it interested you? Was it the mathematics?
D. Geselowitz:
Yes. The idea of studying the electrocardiogram appealed to me. The major idea was that the heart generates electricity and the body conducts electricity. Therefore there are the currents that result from the electrical activity of the heart and therefore potentials on the skin. The question, "What is the relation between the surface or skin potentials which can be measured as the electrocardiogram and the sources of electricity in the heart?" This sounded to me like an intriguing problem. As I said, the head of the electromedical laboratory, Herman Schwan, was interested in impedance. There had been a study where Schwan worked with a cardiologist that determined the electrical properties of body tissues with relevance to electrocardiography. This was with the low frequencies below 100 or 1000 hertz. One of the tissues that Schwan studied extensively was blood. The impedance of blood depends on the number of red blood cells, or the hematocrit. Therefore the idea came up that maybe a little handheld device could be used to measure hematocrit based on the impedance so it could be taken to the bedside Blood would be drawn, and put into this device which would produce a reading right then and there that gives the hematocrit. Okada had developed this device and was interested in using this as an entrée into the commercial world.
M. Geselowitz:
Did he then leave Penn and pursue that?
D. Geselowitz:
He pursued it for a while, but that fell through to a large extent. The big advantage to me was that he did not leave immediately. He became my mentor and for a couple of years we shared an office and worked together. Okada introduced me to all the work in electrocardiography. When he finally did leave I took over. He was also my Ph.D. supervisor. Schwan my supervisor – "de jure" I guess you would say – but Okada was de facto, but he was a young assistant professor.
M. Geselowitz:
He was only a couple years ahead of you.
D. Geselowitz:
Right, therefore it was not considered appropriate for him to be my actual thesis supervisor. However for all intents and purposes he was my thesis supervisor. Schwan's name went on the thesis. Working with Bob Okada was a very valuable experience.
M. Geselowitz:
Was there just one cardiologist through all five of these engineers that worked on this project?
D. Geselowitz:
No. The cardiologist primarily involved in this was Calvin Kay who headed cardiology. Then after a year or two the cardiology division brought in a young cardiologist named Stanley Briller. Briller had a specific interest in electrocardiography and had worked with Charles Kossman in New York. Kossman is a cardiologist who had done some important work. Stan Briller came to cardiology and the two of us worked together for many years.
M. Geselowitz:
Was he affiliated with the University of Pennsylvania medical school?
D. Geselowitz:
Correct. The University of Pennsylvania medical school was two or three blocks away from the engineering school, which was great – and somewhat unusual. In many cases the medical school is in a separate city. At Harvard it's across the river. That isn't far, but people have said that can be a big impediment. However at Penn it was just down the block and the hospital was there.
M. Geselowitz:
What specifically was the subject of your dissertation?
D. Geselowitz:
I became really interested in how the potential field that exists on the body surface could be represented in terms of cardiac sources. At any instant in time we have a potential distribution over a surface, and it varies with time during the cardiac cycle. The question was how this could be represented at any instant in time. There was some work along these lines with the idea of the multipole expansion. There had been a lot of work where a lot of the emphasis in electrocardiography was on the heart as a dipole source. This goes back to Willem Einthoven, who really started the field of electrocardiography and made a number of outstanding contributions for which he was awarded the Nobel Prize. He presented the concept that the heart electrically acts approximately as a dipole. A lot of the early work involved this dipole concept and how well the dipole represents this potential distribution on the surface. However to represent it completely so to speak we had to go beyond the dipole. The multipole expansion is a way of expressing an electric potential on a surface and is an infinite series the first term of which is a dipole and then so forth. In an infinite medium the dipole field potential varies inversely with the square of the distance from the dipole. Then there are terms that go off with the cube, and so forth. This is the multipole expansion. I showed how this concept could be used in a bounded volume conductor such as the torso.
M. Geselowitz:
Did that give a better prediction?
Thesis research as a Science Citation Classic
D. Geselowitz:
No, it did not give a better prediction. It was a way of rigorously handling the field mathematically. Incidentally, a number of years after I published my thesis this paper became a Science Citation Classic. I was very proud of that.
M. Geselowitz:
That is based on the number of times it is cited.
D. Geselowitz:
Right. [Dr.] Eugene Garfield created the Science Citation Index. He was in Philadelphia a couple blocks away. He had a brilliant idea. If I may digress, generally what is done is a literature search to find papers in which one is interested. One can go back and trace what happened in a particular subject of interest. Garfield said, "Aha. Let's go forward. If one finds a key paper that was written ten or twenty years ago it would be very valuable to find who cited that paper in subsequent years." That's what the Science Citation Index does. He came up with this scheme, a compilation that enables one to find citations to different papers.
M. Geselowitz:
One can trace the development of that particular field of literature over time.
D. Geselowitz:
Yes, right. Then he started identifying Science Citation Classics based on the number of citations.
Employment and research at the University of Pennsylvania
Decision to accept position
M. Geselowitz:
Great. What happened after you got your Ph.D.?
D. Geselowitz:
Again at that point I had a choice to make. I actually interviewed with a number of companies. In particular I had an offer from Bell Labs. That was very tempting. However I was also offered a position in which I could continue on the faculty of The Moore School, and I opted for that.
M. Geselowitz:
What particularly caused you to make that decision?
D. Geselowitz:
You asked me before about the Ph.D. Sort of from the beginning I said, "Aha. This is what I want to do." I had an interest in being involved in the academic world.
M. Geselowitz:
In teaching?
D. Geselowitz:
Yes, in teaching and continuing to do research.
M. Geselowitz:
In those days Bell Labs was kind of unique in the industrial world in that the researchers there got to do pretty pure research.
D. Geselowitz:
Yes. That was very tempting.
M. Geselowitz:
That is a little different than going into straight industry.
D. Geselowitz:
Yes. My dissertation dealing with the electrocardiogram was no problem or maybe a plus as far as Bell Labs was concerned.
M. Geselowitz:
It would help you think outside the box.
D. Geselowitz:
Yes.
M. Geselowitz:
You had done the signal-to-noise and shielding work, so you understood applications in telecommunications and that sort of thing.
D. Geselowitz:
Right.
M. Geselowitz:
And you had worked in the microwave range.
D. Geselowitz:
Yes.
M. Geselowitz:
You had a very good background. Anyway, you decided to stay at Penn.
Electrocardiogram research: signal domain, electromagnetic theory, antenna theory
D. Geselowitz:
I decided to stay at Penn and continue my research. Bob Okada had been working with Paul Langner, Jr., the medical director of the Provident Mutual Life Insurance Company, who was interested in doing research. Langner had built up a little laboratory in the insurance company where he could do some studies on the electrocardiogram. Bob Okada had worked with him, and when Okada left I fell into this slot and worked with Paul Langner for a number of years. That was a very interesting and valuable experience. The work I did with Langner was more in the signal domain than my thesis, which dealt with field theory where at each instance of time there is a potential on the body surface. At any point on the surface there is a time varying potential or signal. Langner had an interest in the high-frequency components of the signal that are generally not recorded in the conventional electrocardiogram. We used FM tape recorders to record the signal with enough fidelity to get these higher frequency components. The question was whether they were diagnostically significant. I worked on that parallel effort.
M. Geselowitz:
Did the work you were doing with Stan Briller continue?
D. Geselowitz:
Yes. That was the major thing. This other one with Langner took just a couple of hours a week of my time.
M. Geselowitz:
Did that continue to be work on the potential and multipolar expansion?
D. Geselowitz:
Yes. One of the things in which I became interested was how to represent the sources in the heart in terms of the electromagnetic theory. I struggled with this for a while before I came up with a way of doing it. I was guided in this by the work of S.A Schelkunov.
M. Geselowitz:
Was he Russian?
D. Geselowitz:
Yes, I'm pretty sure he was Russian. He may have had a connection with Bell Labs. He had written a classic textbook that was used in a course I took on antenna theory. He used the concept of impressed currents; that is, an antenna is energized in some way and this gives rise to impressed currents in the antenna. These current sources can then be used to calculate the field of the antenna. I came up with this in terms of the electrocardiogram. There is an impressed source in the heart, throughout the heart, and the question was how this could be represented in terms of the potentials, the electric fields, in the heart. I came up with this. Again that's an achievement of which I'm proud. It was one of those things that, once I had presented it, it became sort of obvious. It was adopted universally and everybody derives it from a very basic standpoint the way I did, and I am not usually given credit.
M. Geselowitz:
Were you the first one to make this analogy to antennas?
D. Geselowitz:
Right. But aside from that, forgetting about antenna theory, the idea is that impressed currents provide a way of treating sources in the heart.
Diagnostic applications of signals in electrocardiography
M. Geselowitz:
You mentioned commercialization and this insurance doctor who was interested in diagnostic aspects of the signal. You did fundamentally important papers in theory. How often did you see that trickle down into work done by cardiologists?
D. Geselowitz:
There was very little of that in a sense. A number of people, electrical engineers, became interested in looking at the electrocardiogram. This was after the end of the Second World War. There were four major laboratories. I mentioned Ernie Frank at Penn; Otto Schmitt (who incidentally invented the Schmitt trigger) in Minnesota; Carl Burger at Utrecht in the The Netherlands; and Richard McFee at Michigan were applying these basic ideas with the emphasis on the dipole – how the dipole can be calculated from the surface and how well the dipole accounts for the potential. Then there was the next generation, which included Okada and myself. Others then got into this, and what has happened subsequently is the ability to achieve results from the cellular activity in the heart, and with more and more sophisticated models being developed. These models show how the signal propagates through the heart muscle and how it generates potentials on the skin. There has certainly been an impact from that on electrocardiography.
M. Geselowitz:
As an end user layperson it seems to me that electrocardiography can tell me if I am having a heart attack. It can tell me if there has been damage to my heart, but it cannot predict if I am going to have a heart attack.
D. Geselowitz:
Correct.
M. Geselowitz:
That would be the most interesting for me. Then I could go to a doctor and have him say, "We have got to do something because otherwise you are going to have a heart attack within three weeks" based on the actual signals. Obviously we are not there yet.
D. Geselowitz:
Yes. All this work still does not answer that question. What it can potentially do is improve the ability of the cardiologist to determine what the damage is, where it is and its severity. However one has to go into another domain to be able to predict what is going to happen. By and large, that cannot be done though people have made attempts.
M. Geselowitz:
While you were doing that sort of work you were progressing at Penn through the ranks of Assistant Professor and so forth.
D. Geselowitz:
Yes, right.
IRE, IEEE, and Biomedical Engineering Society
M. Geselowitz:
Before all that, let me ask you when you got involved professionally with the IEEE or the IRE, which is one of its predecessor organizations?
D. Geselowitz:
After I graduated and was offered this position in the electromedical laboratories I became sort of the senior person with Herman Schwan. There were not too many of us. Herman Schwan was thinking very seriously of starting a graduate program in biomedical engineering.
M. Geselowitz:
Were there any others at that time?
D. Geselowitz:
No, but there were a couple of other people interested in the same thing. In the IRE there had been an interest in biomedical engineering topics. There were conferences held and publications. The IRE, and subsequently the IEEE, was a key organization in terms of the field of biomedical engineering and presenting papers and coming up with organizations that could advance the field.
M. Geselowitz:
You are now a historian of the IEEE Engineering and Medicine Biology Society, which was preceded by an IRE professional group. When did that professional group form?
D. Geselowitz:
I am not sure of the date. Herman Schwan was a very key person in that. At that point electrical engineering dominated the field, so the IRE and the IEEE became the focus for biomedical engineering. At the meetings where papers were presented there was a real attempt to bring in other people, for instance mechanical engineers, chemical engineers and so forth. They did presentations at these meetings also, but that was not a major effort in mechanical engineering as opposed to what was happening in electrical engineering.
M. Geselowitz:
Meaning that the IRE professional group was really the biomedical engineering group?
D. Geselowitz:
Yes, in the United States. Of course this changed over the years. Mechanical engineers want to do their own thing through the ASME [American Society of Mechanical Engineering.] The IRE definitely played a key role. There was also an attempt later to use that as a basis for forming a society, to use the IRE and the IEEE and this professional group as a basis for forming a biomedical engineering society. However that never materialized. Therefore the Biomedical Engineering Society was created. People like Herman Schwan were really interested in expanding through the IRE or IEEE..
M. Geselowitz:
The IEEE was formed in '63.
D. Geselowitz:
It was probably the IRE. Getting the IRE to say we can admit, for example mechanical engineers, as full members turned out to be difficult.
M. Geselowitz:
Mechanical engineers?
D. Geselowitz:
Yes, or even people who were not engineers.
M. Geselowitz:
Such as cardiologists?
D. Geselowitz:
Yes. That did not pan out.
M. Geselowitz:
I see. Is that why there is a separate Biomedical Engineering Society outside of IEEE?
D. Geselowitz:
Yes.
M. Geselowitz:
Did most biological engineers at this point belong to both?
D. Geselowitz:
Correct.
M. Geselowitz:
Your students presumably belonged to both?
D. Geselowitz:
Yes.
M. Geselowitz:
When that fell through you stayed active in the IRE and the IEEE.
D. Geselowitz:
Yes.
M. Geselowitz:
In fact you edited one of the engineering journals.
D. Geselowitz:
The journal that the IRE was publishing became what was the most important journal in the field for many years and even for many years after the Biomedical Engineering Society was founded. I was editor of the Transactions for a few years.
M. Geselowitz:
Did you also try to publish chemical engineers and others?
D. Geselowitz:
It was open to anything that was bioengineering. Most of the papers were electrical, but there was no attempt to exclude the other fields. As a matter of fact, bringing them in was a real interest. There was a concept that this Society and publication could be the bioengineering focus for the United States.
M. Geselowitz:
The IEEE may have missed an opportunity there.
D. Geselowitz:
That could be, although in retrospect I am not sure whether the other types of engineers would have been happy with it. Yes, I think they did miss an opportunity.
Development of biomedical engineering program at U. of Penn.
D. Geselowitz:
Related in a way were the graduate programs in biomedical engineering. Herman Schwan was very much interested in developing a Ph.D. program so that we could start to train students in biomedical engineering. There was a definite move on his part to do this at Penn and we started teaching courses toward that end. All of this was within The Moore School of Electrical Engineering. Schwan developed a course and I developed a course. They were EE courses that were biomedical.
M. Geselowitz:
Was this at the graduate level?
D. Geselowitz:
This was all at the graduate level. There was then a move to formalize this, and because of the politics with The Moore School that we have talked about at some length, Herman Schwan's idea of creating a graduate Ph.D. degree in biomedical engineering was modified. I think it became biomedical electronic engineering. There were people politically powerful enough in the university to say that The Moore School could not do biomedical engineering.
M. Geselowitz:
The chemical engineers wanted it to be biochemical and the mechanical engineers wanted it to be biomechanical?
D. Geselowitz:
Right.
M. Geselowitz:
The title was a political issue, then.
D. Geselowitz:
The title could not be biomedical engineering. However we did get this biomedical electronic engineering program, which was the first biomedical engineering Ph.D. program in the United States. The students were essentially Ph.D. students. Support was obtained from the National Institutes of Health (NIH). There were four schools that initially received grants from NIH, Penn being one of them, which supported this effort in graduate work in bioengineering. I became involved with this with Herman Schwan, so this occupied some of my time. In the development of this graduate program we had to start from scratch of course.
M. Geselowitz:
Were any of the other three NIH grantees, or the next generation, more successful because of local politics in better integrating biomedical engineering as a single discipline?
D. Geselowitz:
There were unique approaches. I am trying to think of the very early ones. Johns Hopkins was one of the four and they established biomedical engineering as a subdepartment in medicine. That was where it existed for many years.
M. Geselowitz:
Could one get a Ph.D. there?
D. Geselowitz:
This was a Ph.D program.
M. Geselowitz:
This was a Ph.D. program in the medical school?
D. Geselowitz:
In the medical school as a subdepartment. The Ph.D. programs in universities are controlled by the graduate schools and can transcend the departmental structure – but it never happens that way.
M. Geselowitz:
Right.
D. Geselowitz:
Our program was largely within The Moore School of Electrical Engineering.
M. Geselowitz:
What year did that program start?
D. Geselowitz:
I got my Ph.D. in 1958 and this all took place the next two or three years I believe. It was very soon after I got my Ph.D. and became a faculty member as an assistant professor.
M. Geselowitz:
In the early '60s then you were working with Herman Schwan in this graduate program, editing the Transactions and doing the main and secondary lines of research you mentioned earlier.
D. Geselowitz:
Yes.
M. Geselowitz:
What happened next?
D. Geselowitz:
The Ph.D. program became quite successful. We were able to attract outstanding students. We were really the first to offer a Ph.D. in biomedical engineering although it was as biomedical electronic engineering. Because we had this program we were able to attract students who were interested in getting a Ph.D. in biomedical engineering.
M. Geselowitz:
I presume you had to also hire more faculty members.
D. Geselowitz:
That came slowly. Students were required to take the medical school physiology course and they took courses in electrical engineering, and we were developing courses in biomedical engineering. Incidentally one of the early people hired was Bram Noordergraf who was in The Netherlands and was interested in developing models for flow of blood. Although some of his models were electrical analogs, this was not an electrical engineering thing.
M. Geselowitz:
Right. With fluid flow it would be mechanical or chemical or somewhere between.
D. Geselowitz:
Yes. It involves fluid mechanics.
M. Geselowitz:
We have been speaking kind of parochially about the NIH and the initial four programs in the United States. Did Europe, which began the field as you mentioned earlier, have institutions producing doctors of engineering with such specialties?
D. Geselowitz:
H. Burger was a key person. He was actually a biophysicist.
M. Geselowitz:
At Utrecht.
D. Geselowitz:
Yes. And Otto Schmitt in this country was also a biophysicist. There was this work at Utrecht with Burger that involved his interest in electrocardiography as well as some interest in ballistocardiography which is a mechanical aspect of the heartbeat. As the heart pumps blood there is a momentum being generated. There was some interest in whether this could be detected and therefore the strength and so on of the heartbeat. The idea was that a person was put on a table and then, as the heart beat, the table moved. By observing the motion of the table, information could be obtained about how the heart was pumping blood. Burger was interested in this, so at Utrecht he had this effort in ballistocardiography. Ballistocardiography never really went anywhere. It was an interesting idea, but there were a lot of problems with it. There was this interest in The Netherlands certainly, and this expanded and exploded. The Netherlands is a place where there has been a lot of really outstanding work in the field of cardiology. I do not know if this was a purposeful decision on the part of The Netherlands. It may well have been. They are a small country.
M. Geselowitz:
To focus on one area.
D. Geselowitz:
To this day cardiology involves a lot of biomedical engineering. There are half a dozen or more outstanding laboratories in The Netherlands.
M. Geselowitz:
Initially as the field was growing you could rely on foreign graduate students as well as graduate students from the United States.
D. Geselowitz:
Not really, until Noordergraf got here.
M. Geselowitz:
You hired a professor from The Netherlands?
D. Geselowitz:
Right, and he brought a couple of students of his from The Netherlands.
M. Geselowitz:
Was that the first internationalization of the field?
D. Geselowitz:
I suppose, yes.
M. Geselowitz:
Presumably not too many Americans had gone to study in Utrecht.
D. Geselowitz:
Right. There were some Dutch students at that point who studied with Noordergraf as well. Of course that all exploded.
Employment at Pennsylvania State University
Development of biomedical engineering Ph.D. program
M. Geselowitz:
Right. When and how did your decision come to leave Penn?
D. Geselowitz:
There was a decision at Pennsylvania State University, Penn State, to get involved in bioengineering. As a matter of fact there was a council established by the State to look into biomedical engineering and I served on that council. Someone in the Commonwealth of Pennsylvania had this idea and formed this council. We met every six months or so and kicked around ideas of what we could in biomedical engineering. One of the key people in this was George Bugliarello, who became the president of Brooklyn Polytechnic Institute [now Polytechnic University]. He was working in mechanical aspects of blood flow. Penn State decided that they wanted to get involved in biomedical engineering, so they created a budget and they wanted somebody to head it. Someone from Penn State who was involved in that knew about me through this council. They approached me and asked if I would like to head this new program they were creating at Penn State.
M. Geselowitz:
Would that also be a graduate program?
D. Geselowitz:
No, that decision was mine. There was a budget and the money was there. I think there was an idea that we could use this as seed money to support interdisciplinary efforts to get engineers, biologists and physicians together. My decision initially was to create a biomedical engineering graduate program as a way to achieve this.
M. Geselowitz:
You accepted the challenge.
D. Geselowitz:
I accepted. This was in 1971. I accepted the offer from Penn State. I essentially had complete freedom with the small budget that I had, and I made the decision to make a major effort to create a biomedical engineering Ph.D. program.
M. Geselowitz:
Like the one you had been involved with at Penn?
D. Geselowitz:
Right. Penn State had set it up so that roughly half of the budget was in the college of engineering and the other half was in the graduate school where they had an interdisciplinary program of some kind. One of the first things I did was to move in the direction of creating a graduate program.
M. Geselowitz:
How many faculty members did you have to work with this program?
D. Geselowitz:
It was essentially myself. We had a committee to help run the program where we identified people in the university who were involved in something that could be called biomedical engineering. Then I got the dean of engineering to create a position and brought in someone. Then a couple years later I convinced him to create another position. The three of us were the core people in biomedical engineering. We had a number of others who had appointments primarily in either mechanical engineering or chemical engineering. Penn State also had a rather good program in engineering science and there were some people there who were interested. Therefore in addition to our core group of people who actually had appointments in bioengineering at that point, we were working with others who were quite active.
M. Geselowitz:
Was it situation where students had to take various courses at the medical school?
D. Geselowitz:
One of the first things we had to do of course was to make a formal proposal for the Ph.D. program. We had to develop a curriculum. We could work with the life science departments to get that aspect of the material, and we had courses in engineering departments. Then we started to create our own biomedical and bioengineering courses. I did a count a number of years ago, and it was roughly 50/50 whether programs were called bioengineering or biomedical engineering. Both terms are used. Penn State used the term bioengineering. One of the first things I was asked was about the difference between bioengineering and biomedical engineering. It was crazy, but at that time there were some people who said that biomedical engineering is a subset of bioengineering and there were some people who said that bioengineering is subset of biomedical engineering. However there never was a distinction.
M. Geselowitz:
Now with the ability to do genetic manipulation there is an additional problem. Most people think of bioengineering as genetic engineering.
D. Geselowitz:
Right. The term bioengineering has come to include that, and that has generated a lot of confusion on the part of the general public. A lot of people think of bioengineering in terms of gene manipulation.
M. Geselowitz:
Gene splicing and not all the other stuff that it encompasses.
D. Geselowitz:
That is not what the people in the bioengineering departments are actually doing.
M. Geselowitz:
I guess those two fields may converge due to the advances in technology.
D. Geselowitz:
Yes.
M. Geselowitz:
Building nanomolecular-based organic devices and that sort of thing.
D. Geselowitz:
Yes, very definitely.
Research, Miller-Geselowitz Model
M. Geselowitz:
You did a lot of teaching, a lot of administrative work and a lot of program building. How did that impact your research?
D. Geselowitz:
I was still able to maintain my research. That was definitely of interest to me, so I still got grants to continue the research.
M. Geselowitz:
Was that research in the same fundamental work you had been doing?
D. Geselowitz:
Yes. There were a number of things with which I got involved. There had been a paper written by Richard McFee, whom I mentioned before. Working with a student of his, Gerry [Gerhard M.] Baule, he had measured the magnetic field of the heart. This generated a lot of interest in biomagnetism. And I wrote a basic paper on the theory of the magnetic field that would be generated outside of a volume conductor. The fact that the body conducts electricity makes it a conductor and a volume conductor. In addition to the surface potential of the electrocardiogram the electrical sources in the heart are creating a magnetic field. I wrote the basic paper that gives the theory of the relationship of this external magnetic field to the electrical sources in the heart. Another thing in which I became involved was an attempt to come up with a simulation of the electrocardiogram. I developed a basic idea of how this could be done, talking about the impressed currents. A student of mine, Walter Thomas Miller, and I came up with the relationship of how the impressed current sources are related to what is happening in the heart cells. All this electrical activity is happening in the membrane of the cells, so we came up with a formalism that related the electrical sources to the membrane potential of the heart cells. Therefore we could relate the sources to what is occurring as electrical activity spreads through the heart. In this way Tom Miller was able to come up with a model that successfully accounted for the normal electrocardiogram. Miller also introduced a number of pathologies. Since we knew what these pathologies would be doing electrically we could then calculate the surface potentials. This was then dubbed the Miller-Geselowitz Model. It provided a basis for going from what was happening electrically in the heart cells to the electrocardiogram. This concept was taken over by many of the investigators – primarily electrical engineers working in electrocardiology – and the models became more and more sophisticated over the years.
Collaboration with Penn. State Medical School
M. Geselowitz:
You mentioned earlier the advantage Penn had because the medical school was next to the engineering school, but at Penn State they are quite far apart.
D. Geselowitz:
Correct.
M. Geselowitz:
Did you get involved with the medical school and particularly with Penn State's efforts to make an artificial heart?
D. Geselowitz:
Yes, very definitely. The medical school was quite new at the time. In 1971 the medical school was a couple of years old. One of the people they first attracted to join the medical school was Bill [William S.] Pierce. He was a cardiothoracic surgeon and had been working at NIH on an artificial heart. When Pierce came to Penn State his goal was to develop an artificial heart. He came over to the engineering school to present his ideas roughly around 1970 or '71. A couple of people in the college of engineering jumped on this opportunity and worked with Bill Pierce. We developed a small group of engineers who worked on aspects of the artificial heart program. The major source of funding for that over the years was through grants from NIH that Bill Pierce had coming in through the medical school. These grants included the engineering efforts. The development of the artificial heart was therefore a major part of the bioengineering effort at Penn State, and of course it involved the medical school.
M. Geselowitz:
What year did you retire?
D. Geselowitz:
I retired in 1998.
Growth of biomedical engineering programs in the U.S.
M. Geselowitz:
Is the program you founded at Penn State still going strong?
D. Geselowitz:
Yes.
M. Geselowitz:
How large is it now?
D. Geselowitz:
The initial programs in biomedical engineering in the United States were Ph.D. programs. They were graduate programs. Some of them were formal programs. As I said, we had the first one at Penn. Different universities claimed they were doing graduate work in biomedical engineering, but more often than not they did not have the formal program.
M. Geselowitz:
"We will staple the certificate to the back of your diploma that you worked on it with someone in a lab" or something.
D. Geselowitz:
Yes. Whether or not it was a formal program they could identify faculty who were involved in aspects of it and claim that they could put together a curriculum that was graduate level. Then there became a movement to create undergraduate programs, and those started to pop up.
M. Geselowitz:
Who pioneered some of the very early undergraduate programs?
D. Geselowitz:
Let's see. Schools like Case Western Reserve, Duke, Johns Hopkins and Brown started to develop undergraduate programs. Penn State decided to develop an undergraduate program. That was created maybe three years ago, so it is fairly new but it represents a major expansion. Bioengineering became a department, whereas before bioengineering was a graduate program. It became the Department of Bioengineering offering an undergraduate B.S. degree.
Philosophies on bioengineering education and interdisciplinarity
M. Geselowitz:
Are you continuing to teach and advise in your emeritus status?
D. Geselowitz:
Yes, I am trying to keep active. I am planning to offer a short course in bioelectricity next semester. I attend the faculty meetings and seminars. It has become universal now in colleges of engineering in the United States that bioengineering must be included. Essentially every college of engineering now has bioengineering, at least to some degree.
M. Geselowitz:
Do you agree with this?
D. Geselowitz:
Well, you know my philosophy was that bioengineering is a graduate program and that one should get an undergraduate degree in a classical field – mechanical, chemical and electrical – and then do work in bioengineering at the graduate level. That lost out. That is a lost cause.
M. Geselowitz:
You are not convinced you were wrong?
D. Geselowitz:
I am not convinced I was wrong.
M. Geselowitz:
Given this issue we talked about earlier with the IRE, do they try and cut across the traditional disciplines?
D. Geselowitz:
Oh yes.
M. Geselowitz:
True departments of bioengineering in the old days would have chemical engineers, mechanical engineers and electrical engineers teaching undergraduate students this new field of biomedical engineering.
D. Geselowitz:
Correct. There are certain key topics to include, such as for example transport theory, biofluid mechanics, applications electromagnetic field theory, biomechanics, and so on. A core curriculum is developed for these undergraduate students and these core courses go across the spectrum.
M. Geselowitz:
Interesting. It has been a long and interesting development of the field. You were the second generation.
D. Geselowitz:
At Penn. Right.
M. Geselowitz:
Schwan was a pioneer.
D. Geselowitz:
Yes.
M. Geselowitz:
Really he was in that first generation.
D. Geselowitz:
Yes, definitely.
M. Geselowitz:
Nationally and internationally.
D. Geselowitz:
Definitely.
M. Geselowitz:
It has been a fascinating interview. Do you have any final points you would like to make? Did I miss anything that you would like to address?
D. Geselowitz:
I would just recapitulate one thing. You asked me about how I got involved and I told you it took me maybe 15 seconds to make the decision as an electrical engineering Ph.D. student to work on the electrocardiogram. I have never regretted that. The whole aspect of biomedical engineering is fascinating – the application of engineering to medicine and biology. It is a huge field, considering all the various aspects of engineering and all the various aspects of medicine and biology.
M. Geselowitz:
And the human body.
D. Geselowitz:
If you look at it as a matrix with biomedical topics on one axis and engineering sciences along the other, almost every intersection yields interesting problems. It is just a fascinating field.
M. Geselowitz:
Thank you very much.
D. Geselowitz:
You are welcome.
