Computer Vision News - April 2021

thing in your work right now, what would it be? We don’t have enough genetic samples. Not enough data? It ’s data, yes, and that ’s the problem. The problem is the datasets are too small. We need to solve that somehow. How will you get data? I think we need to move in the direction of new technologies where it ’s cheaper to acquire similar information. Cheaper technology like X-ray and ultrasound? Yeah, ultrasound, EEG… Why has that not happened yet? It is moving in the direction of portable technologies. I mean you can’t scan everything, but we could do things hierarchically. There is the UK Biobank, which is this enormous adult dataset. We can build models on that. We can then propagate those slightly smaller baby datasets. Then what we want to do is build informed models, more complex models of what’s going on, into the brain. We want to propagate that information using a combination of novel, portable technologies, apps, wearables, and all sorts of things that I don’t know anything about, to be honest, but I would like to be able to make this more useful. What do you think of that? When I was younger, I wanted to work in banking but my uncle told me, “ Never work in banking. Banking is a bad career!” [ both laugh ] Finally I followed his advice. So if a young student decided today to go into medical imaging and study the brain, what would your advice be? My advice is you must absolutely understand neuroscience. You do your best to understand, as far as possible, neuroscience. That will point you in the right direction. Find out 100 more interviews like this with Women in Computer Vision! 345 Emma Robinson