Krishnakant Saboo is a Postdoctoral Scholar at the University of California, San Francisco. His research leverages machine learning to model and optimize brain  stimulation-based treatment of epilepsy. Krishnakant was named a Schmidt Science Fellow at the conclusion of his Illinois PhD in Electrical and Computer Engineering. We caught up with Krishnakant to chat about his time at Illinois, his experience in applying for the Schmidt Science Fellows program, and where his work is taking him next.

Tell us a little bit about what you did at Illinois:

I was in the Electrical and Computer Engineering program working with Professor Ravishankar Iyer. I had come in with an interest in doing something in healthcare. My background before that was in electrical engineering, signal processing, and so on. And this group turned out to be perfect, because they are engineers, computer scientists and data science modelers at heart. Their work spans a very wide range, including healthcare applications because of the very strong collaborations that Professor Iyer has with various groups at Mayo Clinic. Mayo Clinic is one of the leading research hospitals in the US. A great resource in terms of working with them is that their clinicians are fantastic. They really care about the patients, they care about finding the right set of technological tools that can be developed to help those patients. They have the clinical insights. And they also have the data. So at Illinois, our group could bring experience in machine learning and modeling to the Mayo Clinic’s clinical insights of what the patients need, the data, and also the biological insights, which we don't have. Then it's about how we can work together to create something that neither of us could have by ourselves.

There are several [senior researchers] in the group whose work is in clinical trial or is in the process of being implemented for patient care in the pipelines at Mayo Clinic. That’s something that inspired me and motivated me to get into research. And I was mostly interested in doing something in brain disorder space. So I worked a bit on Alzheimer's disease, a bit on epilepsy, and on how you can use the existing data and the clinical knowledge that the neuroscientists and the physicians have to develop models that can diagnose the disease in advance or predict what might happen with a patient in the future.

Tell me about how you came to be a Schmidt Science Fellow. What’s that process like?

It was a really intense and rewarding process at the same time. I think a couple of years before I applied I had already seen the nomination call for this from my department. A couple of things that really helped me at Illinois were the emails that the ECE department sends out about fellowships, scholarships, awards and so on, and also the weekly mail that comes from the Graduate College about opportunities for students. I would always keep looking through there to see if there was something that I could pursue, and that's how I came to know about the application for the Schmidt Science Fellowship program. It piqued my interest because it was about how you can go and work in a very different area from what you were doing, gain new skills in an inter-disciplinary fashion, and then continue working.

During my doctoral studies at Illinois, I became deeply interested in research. I like teaching. So I thought maybe I should look for academic roles in the future. And the fellowship was a good way of furthering my skills, trying to gain some new expertise, and then trying to apply for positions. That's how I got interested in the fellowship. I think when the application process cycle started for me, I had some groundwork done in terms of ‘what do I want to do next?’ ‘Does it fit with the vision that human science fellows have? And is my background ready for applying for something like this?’

Well, it seems your background was absolutely right for applying to something like this! You mentioned that the Schmidt Fellowship involves branching out into a new opportunity for you. What’s that new challenge?

There are a couple of ways I'm thinking about the new area. One is in terms of addressing this problem of brain disorders. A majority of my work during my PhD was: how do you diagnose the disorders better? How do you improve our understanding of what's happening in the disease, and so on. But to be able to address the problem in totality, you also want a way to treat the conditions. That's something I barely started to scratch the surface of during my PhD, and I wanted to delve a lot more into it. So that's what I'm doing for my postdoc: trying to look at how we develop or optimize brain stimulation treatment for neurological disorders. I'm specifically looking at epilepsy because the infrastructure is well set up in terms of data. They have some understanding of how stimulation might help patients, but then there's also a lot of scope for improving the efficacy of treatment. So that's one way I'm thinking of the pivot. The second is just in terms of the techniques that I will be using to do this sort of work. A lot of the work that I've done before was primarily focused on machine learning methods. Now I want to delve a lot more into the computation models that neuroscientists have developed to see how those sorts of models could be useful here. For me, it's going to be about learning a lot more about the neuroscience, the modeling techniques that neuroscientists have developed, and seeing how I can leverage those, learn those, in a way to address this problem. I think that’s where the placement I’m in at UCSF became [a project] that I really wanted to [join], because they have that data. They’ve been working with those patients. They understand that brain stimulation technology very intimately because they’ve been having some patients for more than 10 years now. 

It's fascinating to me that you’re working on how you can combine these methodologies and ways of thinking and then apply them to really large datasets. You can find insights you were never able to get before because you’re looking at the problem from a multidisciplinary view.

That's the hope, that we are able to put things together in a meaningful enough way that we get new insights about why stimulation is working for patients, but not working for some other patients. And if it's not working for somebody, how do we then change it so that it starts to have some benefit for them in terms of controlling their seizures, or helping them improve their quality of life, and so on.

I’ve got two more questions for you. First: what’s that lab you mentioned?

I’m at the University of California, San Francisco. It’s called the Chang Lab. My PI’s name is Dr. Edward Chang. He’s a neurosurgeon, and there are lots of branches in the group working on different problems. Epilepsy is just one of them.

One final thing: you trained in a discipline, and now your new direction required a pivot—I heard you use that word. That’s interesting to me, because I’ve talked to folks who have had a career change after graduate school, and while it’s thrilling, it takes some courage. What made you want to expand your horizons into interdisciplinary work?

I find this very fascinating. This entire field: how the brain works, how it fails in unimaginable ways, and then trying to figure out what may be going on. So that by itself I find scientifically very interesting. And then I was trying to look at it from one perspective, from a very engineering-focused perspective with machine learning. …I saw that if I had more insights about the science, I could develop machine learning models in a different way than I would if I didn’t know all that science. So that was part of the reason to say, ‘I can push the boundary a bit more for myself in how much I know about the biology and neuroscience.’ So that became a push for the personal joy of doing it. The other factor was always this motivation to do something which could eventually help the patients and doctors. I thought that trying to learn these new techniques might help us address the problem in a way which we’re currently not thinking about. If it works out, great, but if not, then I learned something cool anyway.

 

This interview has been lightly edited for clarity. For more information on the Schmidt Science Fellows program, consult the Graduate College's Fellowship Finder database or contact the Graduate College Office of External Fellowships.