Short bio
Bram van Ginneken was born in Nuenen in 1970. He studied Physics at Eindhoven University of Technology and Utrecht University. In 2001, he obtained his PhD at the Image Sciences Institute on Computer-Aided Diagnosis in Chest Radiography, where he continued to work and set up a research group on medical image analysis. His PhD research on automated detection of tuberculosis resulted in medical device software called CAD4TB. With installations in over 70 countries worldwide, CAD4TB is the most widely used autonomous AI solution for the interpretation of medical images. In 2010, he moved to Radboud University Medical Center where he set up the Diagnostic Image Analysis Group and was appointed full professor in 2012. He (co-)authored over 300 publications in international journals. Since 2010, he also works for the Fraunhofer Institute for Digital Medicine MEVIS in Bremen, Germany. In 2014, he founded Thirona, a company that develops software and services for medical image analysis. He pioneered the concept of challenges in medical image analysis and created grand-challenge.org.
A longer biography
Youth
I grew up in the south of the Netherlands, in Nuenen, a village best known for the fact that Vincent van Gogh was unhappy there for a few years of his life. The second most famous person from Nuenen, dixit Wikipedia, is a computer scientist! Edsger W. Dijkstra lived a few streets away from where I was born. My niece was his neighbor for a while. Dijkstra’s Archive with a lot of handwritten manuscripts (even from after 2000!) is amazing.
After high school at the Augustinianum, I decided to study physics at the nearby Eindhoven University of Technology. I’m not really sure what made me do that. Not my high school physics teacher, for sure. He was a friendly and very patient guy with the tendency ask questions to one his pupils and when this pupil had no idea what to answer, he would stare out of the window for prolonged periods of time. This is how I remember high school: sitting in a classroom for hours on end where nothing much ever happened. Among all the things I learned there, I liked physics and at least I did not disagree with what my teachers taught me about it, as I did with some other teachers who clearly had a different opinion about literature than my sixteen-year old self.
Physics student and journalist
Studying physics was quite interesting, and I liked all the math courses as well. But the endless series of exams you had to pass became boring at some point. You always had to solve a few exercises. They were always comparable. You got some situation, an electrical circuit, or an electron floating in a field, or particles under pressure or even people turning flashlights on or off in trains that moved forward with velocities approaching the speed of light, or just a matrix or a differential equation if it was a math course, and you would apply some formulas, hopefully not get stuck in calculations, and you ended up computing the value of something.
After two years I decided to take a break and do my civil service. At that time, every young Dutch male was still required to serve in the army, unless you refused and claimed to be a conscientious objector. I did that. You had to write a letter explaining your thoughts on the matter and present it in front of a committee. I worked hard on a pompous declaration that ended with the magnificent observation that freedom does not live in the barrels of a warm gun, conveniently borrowed from a song by one of my favorite artists. The committee approved it and I could go and look for a state-sponsored job. I found one at the university newspaper. I really liked to be a journalist. You learn to write a lot of stories, have to make sure that are a bit entertaining, they should be more or less correct, and always finished before the deadline. This is was fun and extremely useful. I’d recommend it to anyone who wants to become a scientist, becsue the job of a scientist is remarkably comparable. I continued to work as a free-lance journalist while continuing my studies at Utrecht University.
The courses I really liked during my first years were those about computers. We learned the basics of analog electronics, digital electronics, integrated circuits and CPUs, and we had to program these in machine language and, in a separate course, in Turbo Pascal. Luckily, my father bought me a computer while I was a first year student. I spent days and nights programming the thing, and eventually I wrote a program that played Connect-4 about as good as I did. I did not know that at about the same time, that game was solved by better computer programmers. Next I started to write a chess program, inspired by the thesis of Jaap van den Herik, the first PhD thesis I read in my life. It turned out to be a bit too complex for me to write a chess program that I could not easily beat, but it certainly hooked me on programming computers to do ‘intelligent’ things.
By the time I was ready to do my final research project (equivalent to what you now have to do for a Master’s Thesis), I had more or less decided that I’d get this over with quickly and I’d become a journalist. But then I started to investigate bricks. My research project involved the investigation of how materials reflect light. The idea was pretty simple: shoot light at pieces of bricks under various angles of illumination and measure how much light is reflected back in all possible directions. But can you model that? Can you predict things about the physical properties of the brick if you measure those reflection properties? I liked the fact it was such a concrete and visible topic. For some reason, everything with very small particles like electrons or very big invisible things like black holes, that should be the stuff that gets physicists all excited, never really worked for me. But the bricks did. I wrote a ray-tracer to do computer simulations of my reflection functions, automated the experimental set-up so I could do measurements day-and-night. I almost burned down the lab because the lamp in my now automated and 24/7 running set-up became very hot and the cardboard around the lamp to make sure the light wente only in one direction also became very hot. One morning I arrived in my dark lab room and the smell was very funny and the cardboard was nearly nrunt away completely. I developed a new model using pieces of old models I read about in really old papers (papers that were written well before I was born, that you had to search in the damp basement of the Physics building). I liked doing research much more than I expected! Much of this had to do with the fact that the people I worked with (my supervisor, Maria Stavridi, a Greek postdoc, Astrid Kappers and Jan Koenderink) were genuinely interested in my work. This was very different from what I had experienced as a student so far; to most people working at a university students seemed to be creatures you subject to exercises, problems, and exams, and that you order to carry out so-called experiments which have been done already thousands of times by previous generations of students. I wrote my first paper about my model of how rough objects reflect light. I never worked harder on a paper than on that first one, carefully crafting every sentence. I thought it was brilliant, especially the confluent hypergeometric function that Mathematica told me I needed to normalize a distribution. A few years later someone from Canada sent me a long e-mail with all sorts of difficult questions about the paper. We mailed back and forth and I noticed he understood all of this much better than I did. Eventually he explained me how I could have done the mathematics in much simpler way. That cool confluent hypergeometric function was not needed at all. Important lessons about science that I’ve learnt later on many times: papers are full of needlessly complex stuff and scientist are usually extremely helpful and interested in each other’s work.
I got my Physics degree and was offered to stay for one year at Physics of Man, continuing the work on reflection properties of materials. Shree Nayar, a professor of Columbia University who had just published a paper in Science about a similar reflection model as the one I had proposed (so I was very impressed by him), visited and it turened out he had a funded collaboration with my professor, Jan Koenderink, and he proposed to make a large database with many different types of objects photographed under a complete set of illumination and viewing angles. That turned out to be a brilliant idea. I worked hard on programming a fairly crappy robot to orient pieces of wood with different materials glued on it in many directions, and on making a uniform light source together with the technicians of the group. Kristin Dana, a PhD student of Shree Nayar, collected samples of all sorts of stuff and came from New York to Utrecht with suitcases full of samples and an expensive high-quality CCD camera. For a few weeks we did measurements day and night on pieces of crumpled paper, aluminum foil, peacock feathers, good old bricks, lettuce and even a piece of skin from a corpse that we got from the pathology department at the university hospital. This resulted in the Columbia-Utrecht Reflectance and Texture Database. Kristin wrote a paper about it and this is still one of my most cited articles.
Finally I decided that I did not want to be a journalist, at least not for now. I saw an advertisement in the paper for a PhD position on programming a computer to detect tuberculosis in chest radiographs.
PhD student
To be continued … This concludes the first 25 of my life. In the next episode, we’ll cover the subsequent 25 years and see how I ended up being a professor of medical image analysis at Radboud University Medical Center, in Nijmegen, a small town in the east of The Netherlands.