Would you mind introducing yourself and telling me a little bit about your background before Brown?

My name is Mark Homer, and I went to MIT for a Bachelor’s and a Master’s in Mechanical Engineering, but got really interested in operations research, algorithms and modeling of complex systems. After that I worked in the aerospace and tech industries writing control software for bioreactors and Mars landers. It was a lot of fun. I worked with algorithms and software that automatically guided and controlled the spacecraft. I also worked on unmanned drones and then some point mid-career got a chance to work on projects to analyze medical data. We were trying to recognize someone using their EKG, almost like a fingerprint. I wanted to go further so I got introduced to Leigh Hochberg and BrainGate [at Brown] and it seemed like a good fit. During my PhD, I decoded brain signals to try and figure out what people were thinking, so even if they can’t move their arms, their brain signals can indicate that’s what they want to do.

You studied BME at Brown for your PhD. Can you tell me a little bit about that experience and why you chose Brown?

Brown is strong in computer science and applied math, and my interest was to take my understanding of algorithms and artificial intelligence even further, so it seemed like a good place for that. But whenever you do that it’s very important to have a domain and a phenomena you study, so I found the idea of studying brain waves to be very fascinating. That’s what Leigh’s lab was doing, and the lab was doing it with people with brain stem stroke. I had the responsibility to create study protocols so that you could run these experiments with these people. That was very hands-on as well, which was very exciting. It wasn’t just the theoretical computer science, but working with real people. When you connected speakers to the instruments measuring their brain waves you could actually hear the individual neurons fire, so that was very memorable and very meaningful.

What was your favorite part of Brown? What would you have done differently?

One of the things that stands out about Brown is that it’s very supportive, and they have specific programs - there’s one program in BME where you have to present to the rest of your group and then everyone gives you feedback on your presentation skills. What’s nice about it is that it’s not for a grade or anything, and people don’t have to put their names on it. It’s just honest feedback. It’s a clear signal that Brown’s trying to develop you overall as a person, not just as a PhD program trying to get that paper out of you and make sure that you can test well when it comes to your discipline. Brown within it’s DNA and its culture, people are overall very supportive. Busy, but supportive.

How has studying BME affected your career trajectory?

Well, one of the ways that my PhD directly helped was that my work allowed me to become familiar with healthcare data. Another area that it helped was it gave me an even deeper understanding of the math and the different coding techniques. Although as a manager I don’t often do that directly, when I’m managing people it allows me to know enough to direct them. Brown was also very good about improving communication skills and teaching me to take fairly sophisticated technical concepts and when you communicate it, do it in a way to really break it down and make it more personable and therefore more engaging.

What is your advice for engineering students looking to go into biomedical informatics?

I would say - this is specific and nerdy - take a linear algebra course. It really helps to start from day one to get into the quantitative stuff, even if it’s just like building your own webpage or playing with a computer. Some of the best people you see at this have been doing this since high school. Another piece of advice is to really embrace the software development side of it. That means not only knowing how to code but knowing how to archive your code and knowing best practices to handling when there’s error in your code. It’s almost the equivalent of asking a novelist how to write a book. A novice will say, well, first you want to have a good handle on the grammar of the English language. Grammar isn’t really what novel writing is about, but if you aren’t savvy with manipulating the grammar, then you can’t get to the higher stuff. I think it does help to have a particular domain you’re interested in, like machine learning for patient recognition, or machine learning for speech recognition. There are a lot of public datasets - start to play with those first.

You now work in machine learning at Aetna. How did BME prepare you for your current work?
If it didn’t, then what could you have learned about to better prepare yourself?

Having a PhD from Brown in something that’s both quantitative and related to biology, healthcare and medicine opens a lot of doors. I also think that when you go and work in BME, you really start to get a sense for the whole world of biology and medicine. It’s hard to explain but there’s things that I know about anatomy or the progression of diseases that are easier to pick up because I had the biomedical engineering training.

What made you choose industry following your PhD? What do you like or dislike about it?

There are some life choices and opportunities that open up when you go outside of academia. The emphasis in industry is more on results, whereas in academia, because it’s more competitive, it’s more like you try to be the best at creating these amazing insights. As a result you spend a lot of time polishing what you’re doing, versus moving on to the next thing. I think you wind up accomplishing more in industry. In my particular field you have to be where they can pay for server farms and computational machinery. Instead of mentoring students on a research project, I’m mentoring staff who have decades of experience. So you can get more done, but the tradeoff is that what you’re doing is you’re not necessarily breaking new ground in terms of the fundamentals, you’re breaking new ground in terms of how it’s applied, or how to really get something to work. The one analogy I give is academia is like people trying to figure out fluid dynamics and industry is like the Wright brothers getting the airplane to fly. You need both, but they’re looking at different problems in different ways.