How brain-computer interfaces will alter our interactions with reality | Interview with Dr. Bin He

0:00:12.9At first

Jed Macosko: Hi, I’m Dr. Jed Macosko at Academic Influence and Wake Forest University. And today, we have Professor Bin He from Carnegie Mellon University, who is a bio-medical engineer and studies the brain. So I’m really excited to talk to you and about some of the new advances that you and your students have made at Carnegie Mellon.

But first, tell us, when did you actually go from doing your postdoc at Harvard and MIT to being a professor, and was it first at Carnegie Mellon or did you move around a little bit?

Bin He: Yeah, I moved quite a bit, and so when I finished my time in Harvard and MIT, I started my independent research lab at the University of Illinois, Chicago. So at that time, my first project was a continuation of my work in Boston, that’s about the heart, of course.

And after working for a couple of years on the heart, actually, to be honest, I was not very successful [chuckle] in what I do in terms of seeking federal funding, at that time. And this neuro-engineering or the brain research, which is really something I cannot let go, came back. So I started a new project, and again, re-entering into the brain research, and for some reason, and that just was so smooth in both publication and the grand applications, so everything just [chuckle] grew substantially, so which make my lab with a primary focus in neuro-engineering since then.

Jed: Did you worry in those first few years in Chicago that you weren’t going to get enough publications to get tenure and enough grants to get tenure?

Bin: Well, I think I was not alone, [chuckle] ’cause a lot of assistant professor struggle, well, few of them probably very lucky, never worry, but to my knowledge and experience, a lot of my peers and what we’re usually [0:02:21.0] ____ tracking during the tenure track. You are given only six years to establish yourself. It really depends on how you go. So I was able to publish quite a bit of papers, but for some reason, my proposals [chuckle] were not very well-received. So people think... And then later on, after I did learn something, is I did publish a lot on the topic, but I believe that the project I was working on, it does not demonstrate a significant impact to the science engineering or to the larger society. Therefore, I learned to refocus and identify a significant problem, and then later on, everything just let fly.

Jed: That’s good it all came together. That’s wonderful. But...

Bin: Yeah, that’s right.

0:03:15.4Switching halfway through the tenure clock

Jed: Wasn’t it a little nerve-racking when you switched from the heart to the brain halfway through the tenure clock of six years? And did you worry, "Wow, I’ve only got now half the time to show that I can get grant money"?

Bin: Well the whole... [chuckle] That period was certainly was nervous and under tremendous pressure. I didn’t give up on the heart work, and after I’ve been continuing working on the heart, but I started with this brain basically is that my lab like 50/50, working on both. So in that sense, I saw that if the brain project doesn’t fly, I still would keep going with my established heart research, but clearly, apparently that my ideas on this brain research was resonated by reviewers from journals [chuckle] and from the federal government. So then, I switched quite a bit of the focus to the brain from the heart, which was what I wanted to do from as early as a high school student.

Jed: That’s great, I’m so glad you got back into it.

And then did you go straight from Chicago to Pittsburgh or did you have some other places you went?

Bin: No, when I was in Chicago, and so I was working on electrophysiological source imaging problem, and one of the important direction was to try to combine an integrator with a function MRI. And for some reason that in the environment I were, the access to the MRI machine was not very convenient. So for that reason, actually, I was recruited by the University of Minnesota which has one of the best MRI center in the world. So I moved to University of Minnesota and I stayed in Minnesota for 14 years, and until I moved to Carnegie Mellon University.

Jed: That’s great. Well, being from Minneapolis myself, and my father is a professor...

Bin: You are from Minneapolis. Okay. [chuckle] I lived in Saint Paul.

Jed: Yeah, he’s a professor in the [0:05:28.3] ____ department there. I’m glad you spent 14 good years in the University of Minnesota.

Bin: It was outstanding years, and I loved my time there.

Jed: Yeah, and I don’t hold it against you for going to Carnegie Mellon because my grandfather, my father and my brother are all undergraduate alumni from that university, I think it’s a great place. [chuckle]

Bin: Wonderful. Thank you.

0:05:50.3Mind controlled robotic arm

Jed: So what have you been up to since moving to Carnegie Mellon? Have there been some really things that you’re really excited that you and your students have come up with?

Bin: Yes. Absolutely. One of the reasons, as I mentioned, intellectually, one of the reasons for me to move to Minneapolis was really attracted by CMR, their Center for Magnetic Resonance Research, the world class facility and a faculty expertise. And then later on, my work had started to shift slightly from neuroimaging to more and more brain computer interface. And the Carnegie Mellon has probably the best program in machine learning robotics and an extremely strong neuroscience program. So that was one of the major attraction for me.

And since I moved to Carnegie Mellon, I think my group has made major progress in terms of developing the first noninvasive and continuous mind control of robotic arm in the world. And that was published about now, it’s a year and a half ago. But yeah, that was really exciting.

Jed: That must have been so exciting. It’s really like science fiction to be able to control a hand with your mind. That’s just amazing.

So what were some of the big obstacles to making that work?

Bin: Well, the mind control, or sometimes the way more call brain machine interface or brain computer interface, there will be essentially two approach, a lot of successful story is people use the brain implant. So you drill a hole, you put a chip inside the brain, and then you’ll get a much better signal, do a lot of good things, but of course you need surgery. It’s risky and ordinary citizen would not like to put a chip just because you want to do something.

So the approach that I have been focusing on is, we call noninvasive brain computer interface, and if we put an electro-sensor over the scalp, it’s really electro to pick up electrical signal generated by the nervous system. The challenge is because electrodes are far away from your brain, so the signal is much weaker compared with invasive recording, and also, it’s dirtier. It’s very challenging to process the data to get to the true intention of the human subject.

And to come back to your earlier question, that was one of the motivation, because lots of work I do now involve artificial intelligence, machine learning, and in addition to robotics, it’s really that the signal is already there. The question is how you can develop a novel technology to read a very subtle small signals out of the big background noise.

Jed: That must be very difficult, but I’m so excited that it worked. And do you think that it will always share the place with direct chip implants, or do you think your technology will really be shown to be the easier better technology for people?

Bin: Well, in my opinion, I believe we need both, it’s really not in terms of competition per se, and as a invasive brain machine interface will always is gonna have a better quality or signal. And in the patient, if really need this surgical intervention, and they are very sick, and this is gonna make a big a deal to their life, I think that would probably is an important approach. That’s reason one.

Reason two, if you can address, answer lots of scientific questions using that kind of experimental set up, so which has its own merit and it’s very important. And I can see that it is something that is going to keep going. On the other hand, for the general population, for example, you and me, myself or my son, my family members, if my son say, "I want to be able to say, do mind control of the environment," I will probably not consider it an option that you need a surgery, put a chip inside the brain. So that is not a technology per se, but it’s a limitation of in terms of the practicality or application perspective. So noninvasive probably is a natural way to go, but of course, it’s very challenging technically. So there’s pros and cons for both.

Jed: Of course.

So do you see that the invention that you and your lab published a year and a half ago, do you think that it’ll just go to people who don’t have any prosthetic arms or anything that they just want to control, like a drone, or fly an airplane, or drive a car or... How do you see it being used for people who have no medical problems?

Bin: That’s a wonderful question. So for myself and many people working in this field, I think that we were attracted and we’re motivated because we want to help people who are disabled. That’s how myself started this research decades ago. But when I do more and more actually currently, and also I have been working for about 10 years in my lab, is one of the important application we are pursuing is for healthy population. For example, my lab were the first in the world, it was seven years ago now, to demonstrate that healthy humans, we choose a college student, can fly a drone just by thinking about that. That was really the first time. And humans can do that.

And along the same line, you could have mind control video game, you could have mind control smart house, you can mind control many other things. It doesn’t have to be disabled patient. Just like we cannot live without smartphone... Well, probably this is too strong, it will be very inconvenient if we don’t have smartphones these days. The brain computer interface or mind control technology could become something like that. It’s a system technology that just assisting every human being on this Earth. So that was really one of the scientific goal that I have been pushing and I wish to make my contribution.

0:12:44.3Sign off

Jed: That is such an important contribution, and I’m not surprised that you have become one of the most influential engineers in the last 10 years. So really glad to have you on this show today. Thank you, Professor He, for spending some time with us and telling us about your technology.

Bin: Thank you, it’s a pleasure.