Researchers finding more bacteria with Saga’s help

While hunting for bacteria, doctoral student Saga Helgadóttir discovered a way to simplify her research with the aid of artificial intelligence. Now she is sharing her discovery with other researchers to help them detect elusive particles, cells and organisms too.

Model student. Broke grade records at Westman Islands upper secondary school. Successful sportswoman. Played football in Iceland’s U19 team (her home country) as a teenager. At the University of Gothenburg, her career as a doctoral student has skyrocketed. Among other things her first scientific journal article attracted international attention before it had even been published. It also took her to the prestigious Max Planck Institute in Germany, where she got to present her research to a large group of researchers

Saga Helgadóttir may be extraordinarily successful. But when we meet in the lunchroom at the Department of Physics, it’s immediately apparent that she is very human too. She admits outright that she is shy and that she would much prefer to avoid doing an interview of this kind. But she has actively chosen to go against her inclination.
“I am fairly quiet and don’t like attention. It can generate a lot of pressure when you hear about how important it is to be visible and to develop your social networks. But when I feel anxious about doing something, I know that I need to challenge myself,” she says.

Saga has always pushed herself in the direction she wants. She sees itself as a competitive person and she likes to win. Which had some bearing when, unlike her social sciences oriented parents, at an early age she became interested in science.
“I found it easy to learn things and liked to play maths games on the computer. When I had to choose my focus area, I wanted to take on what was considered to be the most difficult subject, and that was physics,” she says. All the progress made in the research in that field also aroused Saga’s interest. Such as that physicists have managed to use minute gold particles in combating cancer cells.  This later inspired her to want to conduct research in biological physics. And when it came time for her Master’s studies, her choice was Gothenburg.

Why did you choose to move here?
“Because I grew up on an Icelandic island, I needed to move somewhere else to get my education. I’ve always wanted to come to Sweden because I lived here for a few years when I was little. Gothenburg is a city of a size that suited me and I felt right at home as soon as I got here. It has everything you need and its’ all so geographically close too. And then we also have an excellent programme here. Unlike what I’ve seen in other places, we work more actively here with applications and practical exercises. This is something that I really appreciate. I understand better when I see how the theory can be translated into practice.”

What is your research about?
“These days, my research is about the application of machine learning to micro particles and organisms. Machine learning is a subcategory of artificial intelligence (AI). The initial idea was that I would study how bacteria interact with particles, but our research team didn’t have any good method for doing this. The traditional approach is to set up rules for set ratios[KA1]  using algorithms. But this doesn’t work very well when you’re working with blurry microscope films with a variety of lighting.”

How did you solve that problem?
“The head of my research team thought we might instead test machine learning with artificial intelligence. Programming is not my strong suit. I had to learn everything from scratch. But after having conducted tests for a few months, we were able to get the computer to distinguish between particles and bacteria. The recipe for success was letting the computer learn itself using simulated images. The results showed that our method was much more accurate than the traditional approach.”

What is unique about your method?
“Its strength lies in the fact that it can be used in everything from general subjects to studying something very specific, because we use simulated images. For example applying this method to bacteria and passive micro particles has been successful. We have also applied machine learning using another method. Here too, we have got some interesting results in studies of genetic diseases. Other researchers have done similar things. But our way is unique and we have made it available to everyone.”

What is the most important thing you have learned at this University?
“My personal growth is probably the main thing that I will take with me from my time as a doctoral student. The experiences I’ve had and how I’ve developed are the most important things. I don’t need to have the biggest network – I can seek out individuals who are interesting to me. Now I understand what people mean when they say that you can do what you want after your doctorate.”

What do you intend to do after your doctorate?
“I don’t think I will stay in academia forever. I try to think strategically and despite everything, only one in ten people gets a permanent position. So I will probably look for opportunities outside the University, with companies in machine learning or medicine.”

Saga Helgadóttir

Age: 28
Studying: Doctorate in Physics
Lives: In Gothenburg with her boyfriend. Grew up on the Westman Islands off the south coast of Iceland.
Leisure activities: “I love challenging myself in different sports and travelling to new places. I’m also passionate about being vegan, and would like this issue to get more space in public debate".