While most PhD students solve complex problems for their dissertation, Nicola Dell figured out how to radically improve people’s lives.

Dell is part an emerging field of computer science that aims to design and build technologies specifically to solve problems in developing countries as well as in underserved communities here in the United States.

Dell grew up in sub-Saharan Africa and her innate knowledge of the kinds of challenges that people face there — no electricity, no Internet and very few other resources — has informed her work.

For her PhD, Dell focused on improving the diagnosis of diseases like Malaria, HIV and Syphilis that plague populations in Africa, where she grew up. The common diagnostic tests for such diseases is to take a blood sample, wait 10 or 20 minutes, and then interpret a series of colored lines that indicate the result. It turns out, however, that local health workers make a lot of mistakes in interpreting results, resulting in incorrect diagnoses.

So, Dell created a system where workers could take a picture of the test using a smartphone. The program then automatically interprets the diagnosis for them and reports it to the Ministry of Health.

“One of the things that really attracted me to Cornell Tech is that they are committed to a non-traditional setup where things like having impact in the real world counts — which is not the case at most big, traditional universities,” Dell says. “And it's an integral part of the work that I do.”

Cornell Tech caught up with her between trips to Africa to learn about what inspires her now:

Cornell Tech: You had a very unusual upbringing and path toward computer science. Tell us a little about the winding road that led you here.

Nicola Dell: I grew up in Zimbabwe and I went to an all-girls school. And as you might imagine, schools in Zimbabwe don't have that many computer science courses. We had one computer studies course, but they only had enough resources to allow about ten students to take this particular course out of my class of 150. They made everybody do an aptitude test and I was lucky enough to be one of the people who was picked. I liked it, so I applied straight out of high school to do a major in computer science in the U.K. I was on the path, and just kept going.

But you didn’t go straight to get your PhD?

After undergraduate, I moved to Cape Town and lived there for two years working on 3D computer graphics films, mostly for kids.

Then, after about two years, I decided that the entertainment industry was not all it was cracked up to be. It was mostly about trying to sell toys. So I moved to South Korea and worked as an elementary school teacher teaching math and science. It opened up my brain to the fact that I like teaching. So I decided to pursue my PhD at University of Washington.

How did you get into your particular field?

When I started my PhD, I did not know that this field of computing for developing countries existed. I went intending to do biomedical imaging and analysis. Then I met my advisor, who was building tools to improve data collection in developing countries. Initially, I was a huge skeptic because the situation in these communities is just so complicated. I thought: ‘The phones are going to get stolen. They’re going to get broken. People will struggle to use them.’ But he convinced me.

What did you work on besides the phone imaging for medical tests?

The main hook for me was combining my prior interest in computer vision with projects that improve the lives of people in low-resource settings.

Computer vision?

Computer vision is when you are taking pictures and videos using a camera and then processing them automatically to interpret what's in the images.

Most organizations that operate in developing countries still use paper for their data collection. They go out with surveys. They go out with paper sheets. They use paper to track how many vaccinations they've done, how many kids have malaria. They end up with rooms and rooms and rooms of paper and those papers are then very hard to analyze. Someone asks, for example, how many babies were born between June and December, and you have to go through thousands of paper forms to answer that question.

It would be more useful if all of the data was in a usable, digital format. But the process of doing manual data entry is error-prone and time-consuming. So most of the time, people just don't do it.

My work allows people in these organizations to take a picture of their papers, and the app will automatically classify fill-in bubbles and checkboxes. This is good, because it allows people to keep using paper. They like paper — they're used to it and it's cheap — but at the same time, it provides an easy way for them to extract the digital data.

Sounds immensely useful. You just started at Cornell, are you working on anything yet?

I have a lot of ideas and I’m planning to work on a few different things.

The areas that I'm most interested in at the moment are in data security and privacy in the context of poor people in developing countries, as well as in underserved communities in New York City.

Over the last decade, technology has suddenly become available in many communities that previously did not have access to devices like smartphones and tablets. Yet many of these people don't have a mental model of what data privacy and security is or why it is important.

Are you teaching any classes yet?

I am teaching the human-computer interaction masters class. I think it's a really fascinating domain to work in because things just move so quickly.

And how will you teach a class in something that moves so quickly?

The technologies change, but the principles behind them are constant. Often you have to actually go to the place that you're trying to study so you can understand the whole context. Do they have access to smartphones? Do they have electricity? You always need to pay attention to the human side — regardless of what the technology is.